Beene WAS a CN Nia ats FOREST PHYSIOGRAPHY ink eos x Be aril kt LN LDS SE on < ‘eoLisury YWJON jo dep au0z 00h of $8 YMUaaLyH 06 WONT G6 SPM OUT OPNITUOT COT — Nun a cme = "A'N "00 “ONS SALVOd ‘77 yoru an 006 SaTIW 40 31V0S uoisay jeaysny 4amo7 [_ uoiday eaisny voivine s <=> BO =u vi QuoZ oI}OI1y ONIYO109 OL SON3SYSS3uY soe ‘ rsomeal POREST PHYSIOGRAPHY PIPGolOGRAPHY: OF bHE UNITED STATES AN De RINCIPEES OF; SOIES IN RELATIONS LO FORESTRY BY ISAIAH BOWMAN, Pu. D. Assistant Professor of Geography in Vale University LIER S I EDL AL OWN. FIRST THOUSAND NEW YORK JOHN WILEY & SONS Lonpon: CHAPMAN & HALL, LIMITED IQII COPYRIGHT, IQIi BY ISAIAH BOWMAN Stanbope [Press F. H. GILSON COMPANY BOSTON, U.S.A. © 014297922 Vio. 4 Lo Eugene Woldemar Gilgard LEADER IN AGROGEOLOGY PREFACE STUDENTS of forestry in the United States are constantly demand:ng a guide to the topography, drainage, soils, and climatic features of the country. The need for such a book is keenly felt, since students in tie professional schools of forestry have little time for the study of origine| sources of material in a subject which is not forestry but the basis of forestry. On the other hand, such general sources as are available are both too brief and too elementary for the somewhat comprehensive requirements of the American forester. In preparing this book I have attempted to steer a middle course be- tween that purely descriptive writing with which the forester is too often satisfied, and that altogether explanatory writing which the techni- cal physiographer is inclined to'regard as the real substance of a scientific book. The descriptive portions are intentionally rather comprehensive, for the relief and not the explanation of it is of immediate value to the forester, no matter how important the explanation may be in assisting him to appreciate and remember the relief. A chief concern has been the reduction of geologic data to a minimum. A geologic statement frequently runs off into so many consequences that the most important of these almost force one to a more extended and complex discussion than the forester, a lay student of geologic and geographic science, can assimi- late. Emphatically, some geologic data are essential, but only in so far as they have an immediate physiographic bearing. A further point concerning the organization of the material of this book requires statement here. It seems so clear that one can not know forestry without knowing under what physical conditions trees grow, that one finds it impossible to see how even the least philosophical view of the subject can exclude a knowledge of physiography. It would seem that one should pay a great deal of attention to lumbering as re- lated to drainage and relief, to silviculture as related to soils, climate, and water supply, and in general that one should emphasize the forester’s dependence upon physical conditions. This would appear to be so plain a doctrine as not to require restatement here, were it not for the fact that some students of forestry and even some schools of forestry still vii Viil PREFACE pay too little attention to the subject. If the forest is accepted merely as a fact, and the chief concern is its immediate and thoughtless exploita- tion, physiography may indeed be the fifth wheel to the coach, although even so practical a view as the lumberman’s must include some knowl- edge of tonography and drainage if merely to put forest products upon the m2-xet. But forestry is more than lumbering, and if forests are to be ¢ aserved, if they are to be improved and extended, every direct rela- tiou of the tree to its physiographic environment is vital. The title, ‘Forest Physiography,”’ does not imply a book on forestry but rather a book on physiography for students of forestry; and, as nearly as has seemed advisable from the nature of the subject, it has been prepared for their special needs. It is hoped, however, that the book may be of service to historians also, and to economists, since a knowledge of the physiography of the United States has heretofore depended upon one or two short and general chapters on the subject, or upon a study of hundreds of original papers and monographs. No attempt has been made to show the connection between soils and agriculture, which is the general theme of text-books on soils. Neither has it been attempted to make a complete classification of all the types of rocks and soils found in nature. The distinctions which such a classi- fication implies may be serviceable to the geologist, the farmer, and the gardener, but they are too finely drawn for the forester, whose needs are met by a broader classification based upon qualities of wider application. It is our purpose to discuss the origin of soil, and the physical and chemi- cal transformations it undergoes in the process of gradual decay and of interaction with the plants that thrive upon it. Soil water, drainage, plant foods in the soil, soil warmth, and soil improvement are additional topics; the most important point of all concerns the actual preserva- tion of soils that occur upon forested lands. In a broad way all soils are an inheritance from a geologic past; they are slow of accumulation, precious, and vital to man’s welfare, for agriculture is the basis of our modern organized life, and soil is the basis of agriculture. Imprudent forest cutting and thoughtless land tillage tend to disturb a natural balance between great forces. That they may be a sowing to the wind is amply shown by the whirlwind of destruction which man is reaping in extensive tracts of deforested, soilless uplands of America, Europe, and Asia. Foray affects not only trees but also soils, one of uinls greatest of man’s geologic inheritances. I am under obligations to many for advice and assistance. First of all to Prof. H. E. Gregory, who has given most generously of his time in reading both manuscript and proof. Prof. J. Barrell gave helpful advice PREFACE 1x in the preparation of certain chapters in Part One, and Prof. J. W. Toumey of the Yale Forest School supplied important criticisms. Dr. G. E. Nichols of the Sheffield Scientific School made a large number of corrections and alterations in the botanical descriptions. Prof. E. W. Hilgard of the University of California, and his colleague, Prof. R. H. Loughridge, have read Part One with great care, and the benefit of their searching criticisms lays me under deep obligations. My obligations to Professor Hilgard extend beyond this, however, for his great work on soils has been an invaluable source of experimental data. The United States Geological Survey has followed its usual generous policy and supplied many of the illustrations. Prof. R. DeC. Ward has kindly allowed me to use the expensive original drawings of two climatic maps after Ké6ppen. A number of publishers, acknowledged in a sepa- rate list, have permitted the reproduction of illustrative material. I have also obtained illustrations from the Canadian Geological Survey, Prof. C. R. Gould of Oklahoma University, Prof. C. A. Reeds of Bryn Mawr, Mr. H. Brigham, Jr., of Cortez, Colorado, and a:‘number of my students in forest physiography. ISAIAH BOWMAN. YALE UNIVERSITY, June Lo, IgIt. ACKNOWLEDGMENT OF ILLUSTRATIONS Fig. Fig. Fig. Fig. Fig. 10, p. 91, from H. W. Wiley, Soils, Chemical Publishing Com- pany, Easton, Penn. II, p. 99, from E. W. Hilgard, Soils, Macmillan Company, N. Y. 130). Lit, and Fig.-14. p..113, tomy hk. De: Ward, Climate, G. P. Putnam’s Sons, N. Y. Based on Koéppen. 20, Pp. 146; Fig. 48, p. 1890; Fig. 61, p. 2305) Fig 02% pew eur from F. H. Newell, Irrigation in the United States. . 80, p. 285, from Gilbert and Brigham, Physical Geography, D. Appleton & Co. 271, p.676. From E. de Martonne, Traité de Geographie Physique, Armand Colin, Paris, France. The largest number of illustrations are from the Collections of the United States Geological Survey. A number of the block diagrams of the Basin Ranges are from W. M. Davis, Mountain Ranges of the Great Basin, Bull. Mus. Comp. Zool. CONTENTS PART. ONE— THE. SOIL CHAPTER PAGE I. THE ImporTANCE, ORIGIN, AND DIVERSITY OF SOILS............. I The Soil in Relation to Life.......... Paco Sea ae ee neater aE I “bhersoiltand) thesBOrest). 0. acecsee. tees ca con bee eee ee I atherViaintenance:of-ar soil Cover 2. . jee secu ts. ee. 3 SoilemalksinpehOrcest, trv crea ot aera oe Oh ene ge ae vi ithe! @ausestotsoleDiversity sas 2. sees 5 oe) Ne ays es ie 22 PHYSICAL REATURESSOPSOOILS tekclive Shr eee Se ee ee 27 Sizerand: Weightrotisolt Particles: .23 . a) tea) las ci tice 27 Borerspacevand Wil this center een to ha armel Wnesee a cu eaty 28 SpecialeAction(o ty layetete eae tls Solera cgi isan SEs aun a 30 Solgand@S ubsoilsry eee sean eee ee cede sckete A S aN 30 DOU EAM TAT ey TAR ne ostv sue n may ANON Gyr tee mamual ose i deaiien 33 VoameSiltandiGlay. sree ene aew te aah ey cenit cea nfl 1s 34 WATER “SUPPL VS OF SOLLSt ngs le oe i cee oet wary hae a etal OL 41 Relation to Plant Growth and Distribution. .............2.... 41 HMormstol Occurrences senna cea earn eel eaten MMT a 44 WERE OLE MPE RAT URI ssiSte natin. asctaver ees ara ere au. Nae Natta) Uae tet 55 iHcologire Relations a: song ween Se asts Peak NaN MeN ON sat ese ilo een 55 Iniluence ‘on Wateron Soil: Memperatures.. 0 2c. 2 eh 56 Soile@emperature andi @hemical Actions. 000 2h eG es. 56) Influence of Slope Exposure, Soil Color, Rainfall, and Vegetation 58 Temperature Variations with Depth 2.2. sie 60 We CHEMICAT ME HATURES GOR “SOLS: suo 4 ily ects ion cease) Ws etgtaaeh 62 RelativesValueior Chemicali@ualitics:. 2 ees: le es ene 62 SolManerals. cis cane oa SR iy aie ci aes Sener aOR atc ene ee 63 Elements olthetSoiltirce tooo ose kee oe Sits Ge Opals 64 Relations:of soil-Hlements to-Plants..= 225.0. oo eee ee 65 Characteristics and Functions of the Principal Soil Elements.... 66 MotalePlantsFood;;Available:Plant Hoody s2.50.. 4.4 54.5 4.n 73 Determination of ;Soilbertilitys 2 .o.ctee. ese hs ee tree 74 Harmful Organic Constituents of the Soil....... PA Rea eR rte cr oe 76 VI. Humus AND THE NITROGEN SUPPLY OF SOILS.......... (ate eens 77 SourceseanduPlantgelationSe-ui-1. 4. seer ere eae TET OreaniceVMia ttenamuthes Sole twee oe ee to eee erate Rei eae 70 SOME uIMUS Maso eke ee ge ae ae Gres Se ee ere 80 Vie SOLES WOROARIDURIEGIONS .: lhtis dics tasnes cijaete sie dle ac eos baton eee 95 Generalg@walitiessiss ote acne Gee ee Me ee ts Saya 95 IAI alliRS Ot Sepia ese ts geen ae oa kG rs ees een ache aoe Rvs en CORR tyra 07 VDE reS Olle CLASSIFICATION soieg-0. en. ees 2 Gee ec seu Fe MOS a Shae ES pe acer nah 102 RurposeroresoilsAmaly sisi ease oa cio ely Sage ae leech cist suse 104 Different Bases of Soil Classifications.:..................-4-- 105 xi xii CONTENTS PART TWO—PHYSIOGRAPHY OF THE UNITED STATES CHAPTER IX. PHyYSIOGRAPHIC REGIONS, CLIMATIC REGIONS, FOREST REGIONS... mniérO du Ctio mage My epee ape atl Me ee aro a Ria rset Phy siographicyRegions ire ripe. Lc oleh ate le) cue cue uey leet tee eae testes Gliimati ARee ions he hook Pes tecren ach veh s decease ese Pc eee Horest: Regions ities Qo ciamgisines, Saseet meio ee eee IX MGCOASTIBRIANGES!) acre jst the oct He ai uid sv sisecsl > eral cytes Sn nae SubdivaSiomse as os 0 2 eo akescth 3 Soh ce eee ee Goast@Ranges‘of,@alitormia ss: *..20, eis. aes ee cae he Klamath dMiountainss.)8 2c vs cele nln sleet ep ene mene emis CoastsRanges' of@recon) 242. soe Sessiceoe cine cee eet OlympiciMountains.a2 he ace Bese hoon eke eres ere Climate 2Soiland!Horestsi 22 vey acl eens ele yee ee eee XT. ‘CASCADE AND: SIERRA NEVADA’ IMIOUNTAINS. <.. 200.75 sc. ce. ces see CascadesVhountarnse oe eee ye eo ee, sh a ae ee ee en @entrali@ascades:s he een be Se ee Soll Glimatetjand Forests te esate ce sae Sierra Nevada Mountains.) 23. ea a eee ee ee NTT PACTRIG? COAST; WALLEY S 3: ci reg sdf sare coeen ene peer eee cee ee ee General: Geography...) carl besos ole ae ee eee Willamette, Cowlitz, and Puget Sound Valleys...........-.... Great Valley. of: Californias. 2. 32) Welle saan a Valley-of Southern: Californias — 2.42.26 4.1 9 ee Soils;otsthe:Pacitic Coast; Valleyst eas 2) cr eee eee XT. CoLuMBia, PLATEAUS AND BLUE, MIOUNTAINS?! op5 et ey ee eee Columbia Pla teausiye cas ee tea ee ao Extent andre imis.-/sh oeacuall cceatcs ize geet casa eee eet eae eae Buried Topography Beneath the Basalt.................. Drainage Effects of the Basalt Floods:.% 52% enh Deformations of the Basalt: Cover: 22202 | a oto eee Coulees' of the: Columbia: Plateaus’. 25.2502 ae Stréam Terraces ies ie aii ieee ete Ne RLM eae Se a eR Climate. Soils'andt Vieretation. cr 2st) arte erase ene Blues Mountains. 5 eee ee Se ist a ae ae ee eRe EV, (GREAT BASTING 5 fecccool coh lege asic hn) ic Sa TE ee Eee ea Arid Region Characteristics; Hydrographic Features........... Salt Lakesiof the Great Basin®. Black Point Monocline; Colorado Plateaus... 2.5)... eee eee 80. Cross-profile of the Grand Canyon of the Colorado River.. 81. Topographic profile in relation to rainfall distribution from southwest to northeast across the three physiographic provinces of Arizona ....... $2. Relief: map of the Henry: Mountains. 25 3.5.0. oon oe ee 83. Timber zones on San. Francisco-Peaks;Arizona... 250. et eee 84. Mesa forest of western yellow pine in the Mogollon Mountains........... 85. Mountain systems and ranges and intermontane trenches, northern Rockies 36. Locationumap ofa part of the northern=Rockies: 922. 22 Gis eee 87. East-west section, showing flat floor and steep bordering slopes of a typical intermontane’ trench ys sos Sei ca nee ee oat ne eee 88. Coeur d’Alene Mountains, looking from above Wardner................. 89. Coeur d’Alene Mountains, looking toward the crest of the range.......... oo: ‘Cabimet*Mountainstdahow eh se el es ne eee a g1. Topographic and structural section across the front ranges in Montana... 92. Map of Great Plains and front ranges, western Montana................ 93. Hogback type of mountains border, Lewis and Clarke National Forest, IVGombar aie Bn oy Nea ea act nie en oma ee eC 94. A part of the Lewis Mountains, western Montana...................... 05.) Mount:Gould)Wewis Range, Montana. ...2- ca. 25: oe ee os 96. A normally eroded mountain mass not affected by glacial erosion. ....... 97. The same mountain mass as in Fig. 96, strongly affected by glaciers which still-occupycitsivalleysiin.4-<.ac cue ee A ee Se 98. The same mountain mass as in Fig. 97, shortly after the glaciers have melted:frombits valleys. ee Vee WI cays aes ct secre ak eco aS 99. Effects of slope exposure on forest distribution, western Montana........ too. Map showing effects of slope exposure on forest distribution, western Mon- CATV Sis ae Pe ec nen wih Vay RHO Silat ia nates on aiuGa USS Ce tame cate Riese el cea cae ge Lor... View southtacross: Salmon: River‘ Canyon... .26. <<. sc sa sc ts toe PAGE 187 189 192 195 197 199 211 215 210 220 220 223 226 2277 230 231 232 234 239 242 247 248 252 257 259 262 266 267 268 LIST OF ILLUSTRATIONS XIX FIGURE PAGE 1o2. Bitterroot Mountains and Clearwater Mountains...................... 321 103. Map of a part of Bitterroot Mountains, Idaho and’Montana............ 325 TOA lapsotithe;centraluRockiesm tea sis at iyel tenet cea el, op Mes 330 Loss Mastandywesty Boulder plateausn. v0 vf ssn .cik oss fou scsa.. donee. 333 TOOrme A OSATOKay RANE Onan cnr nici sm nen) ars re ete eu ates ene 336 eum CALATIM CR EM AITS eessieisish ais eucoeiey at ee necro NE Oa idee ug 338 108. Terrace and escarpment topography, Green River Basin................ 340 109. Hogback topography of inclined beds, southwestern Wyoming........... 341 110. Mountains of the Encampment district, south-central Wyoming... ...... 343 111. Stereogram and cross-section of the Uinta Mountains arch.............. 340 112. Glacier systems of the Uinta Mountains in the Pleistocene.............. 348 113. Section across highest part of Bighorn Mountains...................... 349 114. East side of limestone front ridge of Bighorn Mountains, Wyoming....... 350 TIS Wallvat head! oficirque;’Bighorn’ Mountains: 2.2022 352 HOw Wormer elacienisystems Omthe Bighorms: = 2:0 00 is lae. Sea a. Bee sk 353 Tie MiaprorunessouthernMRockiess.@ vac ss {snes cen cisi es oe en he es Pace 356 118. Generalized east-west section near Boulder, Colorado. .................. 357 11g. Cross folds on eastern border of Colorado Range....................... 358 TOE VEStIS PAalMSMt Peak) wicca ocgite e scymra nee mis orga ines ents Magbents pay tls arts en tenete 360 121. Longitudinal profiles of five prominent ranges in the Rocky Mountain EOWA CO sie ennevate cite ees tags eeu Pe cee NLU TS accel Larne giant a4 363 122. Section on the common border of Great Plains and southern Rockies. .... 364 123. Old mountainous upland of the Georgetown district, Colorado........... 366 124. Topographic profile and distribution of precipitation across the Wasatch andathe:southermROCKIES aay vat etait es ee ce nes cent atin 125. Extent of the former glacier systems in parts of the Park (east) and the J Sawatcha@west)tranges| of Coloradon 2 kes amie tae Seles 371 126. Landslide surface below Red Mountain, near Silverton, Colorado. ....... 375 127. Looking down La Plata Valley from the divide at the head of the valley.. 377 T26-Western summits of La-Plata Mountains: =: . 22.03.5525 ..60.05. bin se 377 TOO MEVOUTItNVAISOMEBTOUP sac vice ae arcls Wie died aie alee SE Aah 379 Tow Panto: Needle Mountains; Colorado:s...6 ive aot eck oboe en 380 NGtee CLOSSISECHONVOL SaMMeUIs Valley.) ve cen oti tthccaele cotta ett le 383 132. Looking eastward from Hunt Springs across the north end of San Luis Walley suniiscse ie ieee cahl Want MMe Meena Sates orractecngee tach eareloe a tage ne 384 133. Mountain ranges of the Trans-Pecos province................2--.-000-- 388 134. ra Cristobal Mountains, New Mexico. 05.3... i ee ee dee 389 135. East-west section across the Trans-Pecos Highlands north of El Paso, Texas 390 136. Fault-block mountain in the Trans-Pecos Highlands, Texas.............. 301 127) CaballosMountains;New: Mexicor: i222 322i Soup cowie eth ele 392 Tse csectionacross Mranklins Mountains 522 2.7ue a cee come Se oes 393 139. East-west section across Franklin Mountains, Texas.................... 303 Tmowelishers beakcand: Watom Mesa i) (a. hone sree eeisler, oo nacie 304 Tree Wastern border Of the ROCMEeSS eS sya. nii etait aie a ee 1eailtls ale deel 305 TAs Nesade Maya, south-central. Colorado.) 7.220 ies ates tewe | 396 MosaVallevaotskio; Grandewis) Paso; Wexas sen totus or aierce1 eos close Dice sieise sete aaa 400 144. San Mateo Mountains, Trans- Pecos Flighlandsrcsr aie tae sce ats ees 402 t4sA. Timber belts, Capitan Mountains, New Mexico. ...............0.... 403 145B. Range and development of tree species in Lincoln National Forest, Trans- PRE COS PRO VAILCE Het goes aearece rs touch irete Grate net on ce aa Maou ates Masks aes h oaeilees 404 Mow, Geoloric map Ob thejmMexas TEGIONS ya) ores site cesses cies h sapere gets asses eee 406 147. Topographic and structural sections across the Great Plains............. . 407 TAS Glacialiteatures; northern-Great Plains... <2 .c:.-e 4s cee ee we wees 412 Mom MetailsiorpadlandsNebraskas .cc..i. 27. lols hi Metincie ceedee i tisislass o elaiel arcs 415 150. Physiographic subdivisions of Texas and eastern New Mexico........... 417 151. Structure of the Tertiary deposits of the High Plains................... 418 152. Typical view of the High Plains of western Kansas..................... 418 LS53u' by pical: border,topography, of the High Plaims.iy i007 2.2.2 is ened. - 422 no4e erosion escarpment ofthe Eugh Plainse.. oe eee ee eese sis alen= os 422 155. Details of form, eastern border of the High Plains, Texas................ 42 156. Precipitation in the Texas PEPION sea eRe erat eee edeteiuanan Se vesca is Mie deeds 424, XX LIST OF ILLUSTRATIONS FIGURE PAGE TG Emcee etation. Ofsthemlexas TECtOnSsesm raises so)s)0n eureseel Wee sii) so ieee 426 158. Llano Estacado, Edwards Plateau, and adjacent territory............... 431 TO SUMMitsouthe ampasas Plain Wexas... omy... 0 acl n\n (ter uA 160. Summits of the Callahan Divide on the Great Plains of Texas........... 433 161. Diagrammatic representation of a divide of the Lampasas Cut Plain...... 435 162. Edwards Plateau, Balcones Fault Zone, Black Prairie................... 435 103) wscarpment timber ofthe Edwards Plateaus. <5 ccc. 5 42 -cc er eeee 437 164. Ideal east-west section across the Black Hills................6......0.. 440 LOSER Vestermislope On Blache biti sie ace te teria yi ere cacao ae seine eectepttets 441 MOOR WD eviles te OWeLeeres neve ites calito eed) eros syei cue lever eeae leutae nt eat eee em eee 444 TO Vlaprorthellittle BeltiMioumtainsesc-s: cs nee (ten. es yee peepee 447 168. Map of the Highwood Mountains, Montana.....................-..25. 450 T6923 Lhe! Ozarkiregion-and surrounding province. 2... 2is:. qc canes occ 452 170s Choposraphy otethei@zatk replon..) Se sae eee: sce ei 453 iP Sectionsacross the Ozark regions. 2. 0c ae ee ee ee 454 172. Mixed hardwoods, etc., in typical relation to topography, Arbuckle Moun- taitiss Okla ommaees cera cscy serene ci aks Win yatta cnet te Dennen eae Ce AS 173. Border topography, Wichita Mountains, Oklahoma.................... 458 T7Ap wey WICAlevaeweOrPralele uke lanty Seis reir: sep etts ate taste ee asco ee ae 461 175. North-south section of Prairie Plains near Tishomingo, Oklahoma....... 464 7 Os Centerssolice accumulatiomess: * Ns out sarees yee eee esate ae ee 465 177 HOUT. drirtesMeets Of WASCOMSIM 2.22 aoc ret a Se ceee Se pee ene 470 178. Distribution of glacial moraines and direction of ice movement in southern Michigan:and northern Ohio and Indiana.) <=... 2.264.352 471 779: Glacial map: of northern “Mlinois:.< . 12.5 snc Pe 472 £80: Wisconsiniice lobes:about:the Driftless Area. y.2).< = 20.0 ceeseens on eee 473 181. Relations of the drift sheets of lowa and Northern Illinois............... 473 182. Southern limit of the Pleistocene ice sheet and distribution of moraines of the Dakota glacial lobe, North and South Dakota................. 474 183. Old and new channels of the Mississippi at the upper rapids............. 475 ¥64.- Profile across Ibake(Michigan..<:. co. 2 sescscvie one ocr. pela eeeee 476 185. Drainage history of the southern Great Lake district................ 479, 480 186. Superior ice lake and glacial marginal Lake Duluth.................... 482 187. Lake Duluth at its greatest extent and the contemporary ice border...... 482 188. Isobasic map of the Algonkian and Iroquois beaches... ................. 484 189. Map of extinct Lake Agassiz and other glacial lakes.................... 484 190. Distribution of prairie and woodland in Illinois........................ 490 TOT: “Cross bimbers Ol Mexass etre setts sie icis sie ties ioe ee eae See 493 TO2 DO rittlessy Areas Ol: WASCOMSIM ss syne nace. ayant cle se haat care pentane 490 103., Diagrammatic:section in the: Drittless Areas... <2.) : one o eee 490 ToA=. Diagrammatic section ofdViarthas Vineyatdwic 2 fess ce eee ee 506 195. Relative positions of ice during the two stages of the Wisconsin glaciation... 507 196. Section showing the relation of outwash to terminal moraine............ 508 ro7.— Glacialcoutwash topography, Lone Island. .2 22 cn. ee ee 509 TOS. Cross sechioniolone sland snes oe ee le ee eee ee 509 i199. Terminal moraines, soils, and vegetation of Long Island................ 511 200. Characteristic growth of pitch pine and scrub oak, eastern Long Island... 512 201. Effects of repeated fires on soil and vegetation, Long Island............. 512 202. Typical growth of hardwood on the clayey portions of the Harbor Hill moraine; Wongilslandsr.8 ese ass cess se ee ee 513 203. Scattered growth of pitch pine and scrub oak on the sandy portion of the Ronkonkama moraine south of Riverhead, Long Island. ............ 513 204. Sand reef, salt marsh, and coastal plain upland, coast of New Jersey..... 515 205. Swampy divides in eastern Maryland between the Chesapeake and the Atlamtic 2g) os Se ee Aas ee 7 oe ace 517 200) «Cypress trees.of the DismalsSwampe- 0208.6 oon ee 518 207. Albemarle and Pamlico Sounds, east coast of North Carolina............ 519 208. Chesapeake Bay and Delaware Bay and the principal bays tributary to Ghvenriee erences, vente eras nik ow) ooh (cone Ace ea ne 519 209. Finger-like extensions of the Mississippi delta.....................000- 525 210. The lower alluvial valley of the Mississippi.. ........e00cccevecccevcees 527 LIST OF ILLUSTRATIONS FIGURE 211. 212. Pater. 214. 215. 210. 217. 218. 219. 220. 221. 222. 223. 224. 225. 220. DOT 228. 220. 230. 221, 232 222% 234. Agia 235a. 236. Dai 238. 230. 240. 241. 242. 243. 244. 245. 240. 247. 248. Coastal features of Texas, long, simple sand reefs enclosing narrow lagoons Prominent topographic features of the Gulf CoastaliPlain |.) WM niiien(: Cross section of the Gulf Coastal Plain in Louisiana and southern Arkansas One of the timber jams composing the great Red River raft............. akesonthe Rede River. valley sinviouisiamasn sys ee. ae) eeiye a ies AMimberdeadencd-initemporary: rattlakewe kee ee es) OnevofithemWedyRaversratitcd eames wh iiae sina anche WaMu Meek (Si Te Map showing diversion of Red River below Alexandria, La.............. Growth and drainage of the raft lakes at the Arkansas-Louisiana state line rincipalMlakesvor Ori Gantt etre och kel wt sev MMU CRN Uy Olek NK! SwampplandsvonthenWimitedy States sami elon ye lance ey aul TMS acai Rock types and boundaries, Laurentian plate as NPE NURI AN se MMAR Ses Aad dibedbaurentian: Plateaus. Ji 5 tml keke i calalao tele ncun wave iarstie a2 Dea of es in a portion of the Laurentian Pisteans orgy Wiew onthe shore or MakeSt.John; Quebec: sp ee es Ne ts Distribution of the dominant conifers in Canada and eastern United States Typical drainage irregularities in the Lake Superior Highlands........... iBoundariessomthe Superiore ehlamdss: <2) en) salen atl oun Nei tae Deformation of strata near Porcupine Mountain, northern Michigan..... Structure and topography of the southern border of the Superior Highlands Character and relations of the pre-Cambrian and Cretaceous peneplains in IMOGEWELN WASCOMSHMS eee amy Nels Rl chal tte Ni AIS Bel ae BAUR SS MA aay Oana Plateau-like western portion of the Adirondack Mountains. ............. Rectangular pattern of relief and drainage lines in fault-block mountains ofsthereastern Adirondacks ieee eter ach a Mea Wal. Aauiae ae Ar ALU Racy cl Dramacemap ofathe Appalachianimecion wus js .2 ae stale iy ieee ae Structural relations of the various parts of the Appalachian System. ...... Axesor deformation, southern: Appalachians | (22 32 ee ee Phystocraphic:map, of southern Appalachians.) cs ene ye Curve illustrating the relation of topographic relief to lithologic compo- sition in the southern part of the Appalachian System.............. Probable preglacial drainage of western Pennsylvania... ................ Miaximumestace or Wakesbassalce si t/a: t Wie als tonecn inten. neat hos bacarba Pisgah Mountains from Eagles Nest near Waynesville, N. C............. Geologic structure of the Appalachian region................--.0--0005 TRforetall- JM Koybu an ersih ofA] Brey ata ecctst ote lee agen ema g ste ama ren cPRIa Tes Vane UR PEL avesAshevallesBasimy era cc Ue MMe ues ls SUS AIC teaver UUM Re Lars ude Distribution of forests and cleared land, southern Appalachians......... Grassy ‘‘bald” and border of spruce forest, White Top Mountain, Virginia iBrotectionjagaimst crosion by parallel ditches. 2.4 ety ets cel . Erosion checked by covering galleys with brush, Longcreek, Virginia. .... . Erosion checked by brush dams, Walnut Run, N. C.....-..........-... HoPlateauandvescanpment of the BluesRidge: viel y scien eh eee . Blue Ridge, Catoctin Mountain, and Bull Run Mountain in Virginia. ... . 2a. Cross section of the Catoctin Belt, western border of Virginia........... . Cross section of the Catoctin Belt, western part of Virginia.............. . Local development of Triassic rock in the older Appalachians... ......... . Relations of the igneous rocks to the sedimentary strata (Newark) in New A OrSeryrqrena eels ncaa mie any, SAU} cee gn RMT so RM sy cic EATS ae A i te Abalisacdestot themed som eat) Senda ta suri aeue nie AUN NaN sd Maal Naya MOE Is . The four crystalline prongs of the older Appalachians................... . Terminal moraine and direction of ice movement in the vicinity of New BYAO Te Mette neee age AN ae OU eA SUNN Ay Mee AS et RUN ai tea nT INR Renan LEME Ia 3), Characteristic terminal-moraine: topography. (0). 2 yal ai idee ene eae vrouleracrossrcentraluNews Hmelamad yi Nek as ie eo eis se diene alenaiullaiel SR Sectionsouth ofeB lmerelalltMMraimer mya o ne hostess 2 algae ute eda) fell Keri. . Forest growth on a steep and rocky New England hillside, Jamaica Plain, XXil LIST OF ILLUSTRATIONS FIGURE PAGE 264. The geologic and physiographic history of the Connecticut Valley lowland and adjacent portions of the bordering uplands — A, B, C, D... ...655, 656 265. Displacement of trap ridges near northern end of West Rock Ridge...... 657 266. Inferred Cretaceous overlap on the southern shore of Connecticut. ....... 658 207. he NorthvHavenjsand plainsor“ desert’) .07...- 2 fc. ease ee 661 208. Relief map of the central part of the Appalachian System. .............. 665 209. The half-cigar-shaped mountains developed on the hard rocks and the arches formed! by.the bedssof an anticline: 2:27.00. vance erase 674 270. The canoe-shaped ridges of hard rocks and the arches formed by the beds Obtaisyanelim Geis oes el ols eva con vs vce tesla ra Ue elon: si aia cts Seg ae eee 674 271. The development of anticlinal valleys and synclinal mountains from an original consequent drainage in a region with Appalachian structure... 676 272. Varying positions of the plane of base-leveling to hard and soft strata and their relation to anticlinal and synclinal mountains,................ 677 273. Cross section from the Hudson Valley across the Rensselaer Plateau and the @aconicRang ess) lesa ise oecaieeaal seeder sansa Sere oer rare ae eee 680 274. Geologic and physiographic map of the Taconic region..............-... 680 275. North-south section across the northern edge of the Appalachian Plateaus 686 276. Warped surface of the early Tertiary (Harrisburg) peneplain of the central Ap patlachurains) srrcicen., ois avi iss hg kes he Bache te are o ete ee ete cee eee 688 277. Section illustrating the terraces of the Ohio Valley..................... 689 278. Present and pre-Pleistocene courses of Monongahela and Youghiogheny TEVICTS She satis tue eaot ore olsseey elas sets, nes (onto eed sradsmass ons eizasen fe once Ree a7 Seas eC Ie 690 279. Distribution of morainal deposits and direction of ice movement in western ING WHY Orkin Merete EMR SI Ae ea An eel eae ne eee 693 280. Maturely dissected Allegheny Plateau in West Virginia................. 604 281. Map of region between Cumberland Plateau and Highland River........ 6909 232.. Northern: portion,or the. Cincinnatiarchei. e.. ae ce es 702 283. Section across the Nashville Basin of Tennessee and the country adjacent 704 264: ‘Physiopraphic belts.im-central New: Yorker oe. 5:5. s)he ees 708 265. Mapioh portion:ofi News Vonks i. Wire tet ae al a Sette eae See 710 286, 287, 288. Proglacial lakes, Finger Lake district, New York............. 712, 713 289. Channels and deltas of a part of the ice-border drainage between Leroy and Bish ers New MOK ke iin vee sat arc steeape esc esc ak eo sem tt I a 714 290. Gulf channel, looking southeast (downstream) near mouth of channel. 75, 291. Roc drumlins or drumloids:... 2s ee 717 292: ‘Lopographic:types).central New York... so 802 ee 718 LIST. OF PEATRES Frontispiece. ZONE Map or NortH AMERICA. PraTeE J. Crrmatic AND Lire PROVINCES OF NorRTH AMERICA. Pirate II. CoLtorapo PLATEAUS. PLATE III. SouTHERN APPALACHIANS. PLATE IV. PHystoGRAPpHic Map or THE UNITED STATES. PLATE V. GroLtocic Map or Norra AMERICA. FOREST PHYSIOGRAPHY PART: ONE THE SOIL CHAPTER I THE IMPORTANCE, ORIGIN, AND DIVERSITY OF SOILS THE SoIL IN RELATION TO LIFE MEN whose work brings them into touch with the soil and its relation to life do not employ the phrase “mother earth”? in a casual sense. The great hosts of plant and animal life that people the lands in large part have their origin in or draw their support from the cover of land waste whose upper layers are the soil. They are, directly or indirectly, the dependent children of the earth. Viewed from such a standpoint the soil is not mere dirt, a substance to be despised, a synonym for filth, but a great storehouse of energy, a great home, a bountiful mother. Countless billions of micro-organisms — the bacterial flora — throng its dark passageways while the roots of countless higher plants ramify through it in eager quest for food and water. Only less numerous are the earthworms, insects, and burrowing animals that delve into it for food as well as for shelter. To supply all these needs is no mean func- tion; it is probable that no other planet in our solar system has so large an endowment of life-giving, life-supporting soil; the evolution of the life of the earth would have been on far lower levels if the endowment had not been so generous. THE SOIL AND THE FOREST From the standpoint of the forest the soil is a factor of great impor- tance. The home of the tree is the soil and the air; and a forester, whose chief concern is the tree, requires a somewhat comprehensive knowledge of these two elements of the environment of every forest. Without I 2 FOREST PHYSIOGRAPHY soil in some amount tree growth of any kind is impossible, although the amount required to produce a poor growth may be very small. Low forms of vegetation,.such as lichens, mosses, and shruhs of many varieties, may find life possible in a region where there appears to be practically no soil; but a careful examination will usually disclose rock pockets partially filled with small quantities of soil, tiny crevices that contain particles of dust, and joints in greater or lesser number that have caught soil fragments washed from the adjacent surfaces almost as fast as formed. These accumulations afford a foothold for the lower forms of vegetable life which tend to disintegrate both soil and rock and further to increase the amount of available soil. Ordinary weathering will tend toward the same result, and if the climatic conditions, the relief, etc., are favorable, a soil cover capable of sustaining a denser growth of vegetation of a higher order will eventually be formed. In time and through the gradual development of a soil cover a dense forest may grow on what was in a preceding portion of a geologic period a bare rock terrane. While the broad relation of the soil to the forest is thus readily dis- tinguishable, the finer relations are often difficult of determination and there are many physical conditions that evoke no recognizable response in the forest world. A study of the maps, Fig. 23 and Plates IV and V, representing respectively the forest regions of the United States, the physiographic provinces and the geologic formations, will enable the student of forest physiography to appreciate at once that the physio- graphic features and related soil types are of more local development than the broad forest types which they support; and that the finer distinctions between soil types are of little value in understanding the range of a given forest type, however directly they may affect the wel- fare of the individual tree by modifying its habitat. In short, it may be said that the conditions which limit either the growth or the distribu- tion of most forest species are so extreme that they embrace or overlap a large number of physical subdivisions. In general plants are rather impartial as to soil unless the soil char- acters are of an extreme type; relatively few have absolute soil require- ments. Competition among forest trees may drive out some species, in which case the unsuccessful species can not be described as incapable of growth on the soil from which they have been driven; simply, their competitors are able better to use the given resources of soil and air. It often happens that a given species is markedly tolerant of soil and climate except at the limits of its range where competition begets an apparent intolerance. ORIGIN AND DIVERSITY OF SOILS 3 It is concluded that plants possess a peculiar inherent force by the exercise of which they directly adapt themselves to new conditions and become fitted for existence in accordance with new surroundings. Thus plants are thought to have a certain physiologic plasticity or power of self-regulation that tends to adjust them to a new environment, a feature that goes far in explaining the absence of a rigid control of physiographic conditions over forest distributions although an approximate control is often manifested.!. The forester, then, requires a scientific knowledge of soils and climate, but in the final application of his knowledge to the distribution and growth of forests it is often necessary for him to employ somewhat broader generalizations than those employed by the geog- rapher and the botanist for the special purposes of their sciences. THE MAINTENANCE OF A SOIL COVER Everywhere on the land we find at work the two forces of soil making and soil removal. In regions of aggradation the two tend in the same direction; in regions of degradation the soil may be removed as fast as formed and bare rock everywhere exposed at the surface; or there may be established so delicate a balance between soil formation and soil removal that though the soil cover is: continually wasted the process of soil formation takes place at an equivalent rate and a covering of soil is perpetually maintained. The matter of soil formation is of special importance to the people of North America, for not only is denudation (chiefly glacial) respon- sible for an area of bare, denuded country almost twice as great as that of the continent next in order in this respect, Africa, but denudation is probably proceeding at a faster rate on our continent than on any of the other five. The saving quality of glacial denudation in the past has been its occurrence chiefly in mountain regions of the United States and in the upper boreal and the arctic regions of Canada where extreme climatic conditions would largely offset the advantages of soil.2 A physiographic map of the United States, Plate IV, appears to show that approximately one-fifth to one-sixth of the total area is now undergoing alluviation; everywhere else, no matter what process has been active in the immediate geologic past, the surface is now being eroded. The action is so slow in some places as to be exceeded by the rate of rock decay, and in such cases no fear need be entertained for the safety of 1 See especially, V. M. Spalding, Distribution and Movements of Desert Plants, Carnegie Inst. Pub. No. 113, 1909; also E. Warming, (Ecology of Plants, Ox. ed., 1904, pp. 370-372. 2 For an outline presentation of the balance of soil-making and soil-destroying forces that have produced the main soil types now at the surface of the earth see the table, p. 25. 4 FOREST PHYSIOGRAPHY the soil; in other localities the action is rapid and disastrous and its checking should be a matter of the gravest concern. As early as 1890, Shaler! estimated that the soils of about 4000 square miles of country had been impoverished through wasteful agricultural methods, repre- senting a loss of food resources sufficient to support a million people; and that at least 5% of the soils which at one time proved fertile under tillage “are now unfit to produce anything more valuable than scanty pasturage.”’ To us of a later generation this figure appears gratifyingly small beside the figure that would express the deplorable ruin of the past quarter century of reckless timber cutting and baneful neglect of fields abandoned by the upland farmer. The forest is an important factor in soil erosion because. it plays a considerable part in the flow of water by which such erosion is effected. The inequalities of the forest floor offer many mechanical obstacles to the flow of surface waters. Innumerable pools of water collect in hollows and are gradually absorbed by the underlying soil instead of running off at the surface. The leaf canopy catches and reévaporates about 12% of the rainfall, while 10% of it runs along the tree trunks and reaches the ground by a circuitous course. The forest litter, the moss-covered and leaf-strewn ground, is capable of absorbing water at the rate of from 40,000,000 to 50,000,000 cubic feet per square mile in 10 minutes, — water whose progress is delayed by some 12 to | 15 hours after the first effects of a heavy freshet have passed.” While the forest thus plays an important part in the maintenance of a soil cover and in the better equalization of the run-off of streams, it would be a mistake to assume that it is the only agent which accom- plishes these highly beneficial results. It is scarcely more necessary to know that deforestation may permit a precious soil cover to be wasted than it is to understand that many other types of vegetal covering besides the forest effect these desirable results. Of the same order of importance is the fact that the effects that follow upon deforestation are not equally harmful upon all types of topography and soil. If these con- clusions are true it is necessary that the soil, the topography, and the secondary vegetative forces of a given region be evaluated before the statement is made that excessive soil erosion is the necessary correlative of deforestation. That the soil is protected by many vegetal types other than the forest is now a well-established fact. The high alpine meadows of the Pacific 1 N.S. Shaler, The Origin and Nature of Soils, 12th Ann. Rept. U. S. Geol. Surv., pt. 1, 1890-1891, P. 333. 2 B. E. Fernow, Relation of Forest to Water Supplies, in Forest Influences, Bull. U. S. Dept. Agri., Forestry Division, No. 7, 1902, p. 158. ORIGIN AND DIVERSITY OF SOILS 5 Cordillera consist in many cases of a thick turf which supports natural grasses of luxuriant growth. The interlacing roots of the grasses in many situations bind the soil past all reasonable possibility of excessive erosion, the stems of the grasses impede the run-off by breaking up in one locality any incipient streams formed in another locality exceptionally favorable to concentration of run-off, the whole grass cover breaks the force of the falling rain and prevents erosion. Added to these is the influence of ponds and lakes of glacial origin in the higher situations. The retention of the soil above timber line on the Beartooth Plateau, southwestern Montana, is attributed to such a combination of grass cover and stor- age basins;! it was as natural a result that gullying should follow over- grazing of these meadows as that evil results should follow deforestation in the well-known case of the Southern Appalachians. The binding of the soil and the checking of erosion are also effected by brush and vines which in moist regions may spring up in a few years following deforestation and form an almost impenetrable covering. Such a tangle of vegetation offers even greater resistance to the surface flow of water than does the vegetation of a forest, besides permitting the formation of snowdrifts, one of the most important forms of surface water storage.’ The retention of rainfall by the mosses that cover the hill slopes of the Laurentian Plateau, a feature especially well developed in the Labrador peninsula, diminishes the run-off and equalizes it to a degree far exceed- ing that of the thin spruce forests of the region. Even the steepest slopes are slippery with loose dripping Sphagnum moss, whose effects obviously exceed those of the most porous forest floor. The very existence of a steep hillside bog is in itself proof of an unusually powerful retentive effect of the moss cover upon both soil and run-off. All of these consequences are subject to changes in degree depending upon variations in soil and topography. If the soil is very porous the imbibition of rain water is rapid and run-off and erosion are correspond- ingly lessened; if the soil is compact there is little absorption, and run-off and soil erosion are more active. A hill-and-valley country, one consisting entirely of slopes of strong gradient, such as the well- dissected Allegheny Plateau, has a high percentage of run-off and soil erosion, for almost every drop of water falls upon a slope and begins a downhill movement the moment it strikes the surface. A flat surface like the till plains of central Indiana or the outer part of the coastal 1 J. B. Leiberg, Forest Conditions in the Absaroka Division of the Yellowstone Forest Re- serve, Montana, Prof. Paper U. S. Geol. Surv. No. 29, 1904, pp. 15—I0. 2 J. C. Stevens, Water Powers of the Cascade Range, pt. 1, Southern Washington, Water- Supply Paper U. S. Geol. Surv. No. 253, 1910, p. 16. 6 FOREST PHYSIOGRAPHY plain of South Carolina absorbs a high proportion of the rainfall, and soil erosion is of trifling importance. Combinations of these factors are both numerous and variable. A part of the southern slope of Long Island is a natural prairie unforested even before the coming of the whites. It bears almost no signs of erosion, and such as occur had in most cases a very special origin. The absence of erosional features is not surprising when the low gradient of the plain, 10 feet per mile, and the high porosity of the sand are taken into account. These flat- lying porous sands absorb from 60% to 75% of the rainfall, perhaps as great a value as that found on any other area of equal size in the eastern half of the United States. In New England it has been noted that the quick-growing brush and the special qualities of the glacial soil prevent the undue erosion of de- forested hill slopes in the Berkshires where the relief is so strong that landslides sometimes occur. The pebbles and bowlders of the till con- stantly divert the run-off and lessen its velocity, while the bottoms of ravines sunk into the till are in a measure protected from erosion by a pavement of stones derived from the till. In many cases in western Massachusetts and Vermont and New Hampshire steep mountain slopes “have been several times stripped of their forest growth with little, though doubtless some, injury to the soil,” and “the mountain streams are beautifully clear except immediately after a heavy Tain 22) The large number of rock ledges that occur in this region contribute to the same effect. The soil is thin, and irregularities of the underlying bed-rock assist in holding it in place not only by physically retaining it but also by preventing the streams held upon the projecting rock ledges from expending the whole of their erosive energy upon its loose material. The effect of the forest upon the run-off alone is extremely difficult of determination, for soil and topography are in this respect of much greater importance. That forests tend to conserve the run-off is clear; their effects in individual cases, however, may be so small as compared with the effects of soil and topography as to be overshadowed by the latter. “Donner und Blitzen River, in central Oregon, is a very uniform stream with a well-main- tained summer flow, but its area does not produce a tree, except here and there a juniper. On the other hand, Silvies River, which exists under the same climatic conditions as Donner und Blitzen River and discharges its waters into the same lake, is anything but uniform in its flow, although its drainage area is heavily forested. Niobrara and Loup rivers, in Nebraska, are very uniform in flow, but there is hardly enough timber on both areas to build a cabin. Nearly 1 H. F. Cleland, The Effects of Deforestation in New England, Science, n. s., vol. 32, IQIO, pp. 82-83. ORIGIN AND DIVERSITY OF SOILS 7 all the streams of western Oregon and Washington are subject to enormous floods, and all run comparatively low in summer, yet no streams in the world have more densely forested drainage ”) areas. 1 The conclusion that forests are not a guaranty of uniform stream flow, in spite of the fact that they tend in the direction of uniformity, does not diminish the interest of students in such influence as forests do exert, since theirs is a controllable influence. Man can not greatly modify the porosity of the soil or the slopes of the land, and the effects that follow upon these causes are therefore irremediable; but man may save a forest or plant one and thus mitigate effects which he can not wholly prevent. In precisely those regions where run-off. and soil erosion are most ex- treme through unfavorable topographic and soil conditions, man: may find it possible to preserve a tolerable state of affairs by saving the forest from destruction. In regions where the conditions are critical the destruction of the forest may mean the quick destruction of the soil, its preservation the preservation of the soil and the indefinite occupa- tion of the region by man. The retaining influence of the forest on the soil is most strikingly exhibited where the balance between soil formation and soil removal is delicately established and may be easily destroyed. An extreme in- stance is Kanab Creek, Utah, where the burning of the forest and the overgrazing of the pastures have resulted in torrent conditions. The tributaries have become deep washes, many new and deep gulches have been formed, dams and bridges have been destroyed by the floods and coarse gravel deposited on formerly arable valley lands.* It is of importance, then, to examine at the outset the relations of soil denudation and soil accumulation, that we may be the better prepared to study those forces which tend to bind and partially to retain the covering of soil; not only that forests themselves may be perpetuated, but also that the flood-plain soils on the borders of the forests may be adequately preserved and the natural advantages of the waterways retained. SOIL-MAKING ForcEs ? The complex cover of rock waste which we call the soil is the product of a great number and variety of forces. Only the principal ones are here outlined. 1 J. C. Stevens, Water Powers of the Cascade Range, Southern Washington, Water-Supply Paper U. S. Geol. Surv. No. 253, 1910, p. 17. 2 H. S. Graves, The Forest and the Nation, American Forestry, vol. 16, 1910, p. 608. 3 The section on soils is necessarily brief and somewhat technical and assumes on the part of the student a knowledge of ordinary rocks and rock-making minerals as well as an elementary knowledge of chemistry. Those students who are deficient in such knowledge should consult 8 FOREST PHYSIOGRAPHY OXIDATION. CARBONATION. HYDRATION. : SOLUTION. MECHANICAL ACTION OF WATER AND ICE. TEMPERATURE EFFECTS. WIND. BACTERIA. ANIMALS AND THE HIGHER PLANTS. OXIDATION Oxygen is the most active element of the air, and the process of oxi- dation is of great importance in reducing rock masses to soils. The action is perceptibly manifested only in rocks containing iron either as a sulphide, a carbonate, or a silicate. Of these the sulphides are changed to sulphates which are soluble and may be removed in solution. The most common minerals attacked are ferrous carbonate associated with the carbonates of lime and magnesia and the silicates of mica, amphibole, and pyroxene. The minerals become gradually decomposed through oxidation and disintegrate into unrecognizable forms. The oxidation of the iron-making minerals of a rock is always attended by increase in bulk, and when this takes place in cracks and crevices it tends, like the freezing of water, to widen the cracks and to increase the surface exposed to attack. In general the action of the air in soil formation is of secondary importance and depends chiefly on the oxida- tion of the lower to the higher basic forms. The ferrous and ferroso- ferric oxides are converted into ferric oxide or its hydrate limonite, iron rust, which gives to soils containing much iron their characteristic reddish or yellowish colors. The presence of ozone! in air without doubt causes it to have a more active oxidizing effect. Ozone is present in considerable quantity in the air only when the air is free from organic impurities and products of decay. The average amount of ozone in a hundred cubic meters of air is 1.4 mg., but the amount may be doubled after thunderstorms.? Merrill, Rocks, Rock-weathering and Soils, 1897. For purposes of brief inspection of the mineralogical composition of ordinary rocks and the chemical composition of rock-making minerals the tables in Appendix C in this book should be consulted, and a text-book of Lithology such as Pirsson, Rocks and Rock-forming Minerals, t909. 1 Ozone is a very active form of oxygen in which the molecules consist of three atoms of oxygen instead of two atoms as in a molecule of ordinary oxygen. It is formed by silent elec- trical discharges, and is chemically unstable, readily parting with one of its atoms, hence chemi- cally active. 2 J. Hann, Handbook of Climatology, 1903, pp. 80-81. ORIGIN AND DIVERSITY OF SOILS 9 The amount of ozone in the air is determined by the rate of change in an easily oxidized substance — not a very accurate method.! Ebermayer emphasizes the more powerful oxidizing effects of ozone in the air and its formation in the forest in unusual amounts.’ CARBONATION The oxidation of organic materials (both plant and animal remains) by bacteria and oxygen in the zone of weathering produces a concentra- tion of carbon dioxide near the surface. The degree of concentration of this gas in the zone of weathering is appreciated by comparison of the soil air with the atmosphere. The amount of carbon dioxide in the atmosphere is 45 parts In 10,000 by weight; the amount in soil air or soil gases is represented in the following table: AMOUNT OF CARBON DIOXIDE IN SOIL AIR3 Wericetion Parts by weight in I0,000 Air from sandy subsoil of forest........... 38 Air from loamy subsoil of forest. .......... 124 Air from surface soil of forest............. 130 AireirOmMypastunel SOM ae scr. ha cone ss eis 270 Atritrom solericheimy humus;)2. sists sche 543 The carbon dioxide in the soil is the agent in the important weathering process known as carbonation, by which is meant the union of carbonic acid with bases in the formation of carbonates. It is dominantly accom- plished by the substitution of carbonic for silicic acid. To some extent carbonates are also formed (1) by the substitution of carbonic acid for other acids, e.g., phosphoric acid, and (2) by the union of carbon dioxide with oxides not united with other acids, e.g., ferrous oxide in magnetite.‘ The process of carbonation takes place on a vast scale. It is most rapid in the tropics, takes place at a moderate rate in temperate lands, and is least important in the frigid zones and in arid regions. It has a direct relation to the amount of vegetation, since it is chiefly through the decay of the vegetation that carbon dioxide is supplied for the reaction involved in carbonation. On the other hand a soil containing carbon- 1 W. M. Davis, Elementary Meteorology, 1808, p. 5. 2 E. Ebermayer, Lehre der Waldstreu, etc., 1876, p. 202. 3 Boussingault and Levy, quoted by G. P. Merrill, Rocks, Rock-weathering and Soils, 1897, p. 178. 4 C. R. Van Hise, A Treatise on Metamorphism, Mon. U. S. Geol. Surv., vol. 47, 1904, D. 475. ake) FOREST PHYSIOGRAPHY ates is ordinarily fertile and supports an abundant vegetation. The surface vegetation and the soil carbonates are therefore mutually inter- active and helpful. The cumulative effect of the act of carbonation would therefore appear to be constantly increasing amounts of carbon- ate substances. But this tendency is offset or matched by the libera- tion of silica in the process of carbonation; about one and one-third times as much silica is released from the silicates as there is carbon dioxide combined in the carbonates.' HYDRATION The action of hydration is the union of water with chemical com-_ pounds in the production of hydrous minerals. It is the most extensive reaction in the zone of weathering and next to carbonation the most important. It affects practically all of the anhydrous silicate minerals of the igneous, sedimentary, and metamorphic rocks to some degree, and many of them to a great degree. The decomposition products of the rock minerals are almost all strongly hydrated, such as the zeolites, chlorites, and kaolin in the silicate class and aluminum and iron among the oxides. The action of hydration is always accompanied by the liberation of great quantities of heat and by increase in bulk. It is calculated that the transition of a granite rock into arable soil, provided such transition takes place without loss of material, is attended by an increase in bulk of 88%. In rocks as a class hydration effects volume increases which range from a very small per cent to 160% (corundum to gibbsite). In general the increase is less than 50%. Such volume increases prevent complete hydration at any great depth below the surface; partly hydrated rock when artificially exposed at the surface, as in railway cuttings, may become completely hydrated at so rapid a rate as to expand greatly in volume and soon disintegrate. Notable in- crease in bulk does not follow if the pore space is ample; if the pore space is small and the rock dense the action ‘is either incomplete or involves great increase in bulk. Commonly hydration takes place in connection with carbonation and solution. In so far as soil water is consumed in the formation of new (hydrous) minerals it can not be used in the process of solution; the amount so consumed is, however, but a small part of the whole, and solution is therefore a companion process of hydration. 1 C. R. Van Hise, A Treatise on Metamorphism, Mon. U. S. Geol. Surv., vol. 47, 1904, p. 480. ORIGIN AND DIVERSITY OF SOILS IL SOLUTION WATER AS AN AID TO CHEMICAL ACTION Water is the most important weathering agent, not only because of its direct effects but also because its presence conditions all phases of weathering, such as hydration, etc. It has so great a dissolving power that few substances found in rocks are wholly insoluble in it, while in water charged with acids of various sorts many rocks are readily soluble and all are somewhat soluble. The number of such acids is small, but their action is so general that they powerfully aid solution in reducing rocks to soil. Nitric acid is present in some amount in rainfall, in surface waters, and in the soil water, in which it may be supplied in small quantities by the action of bacteria. Sulphuric acid may be derived in somewhat the same manner; an important source in some regions is iron pyrites, which on decomposition may yield free sulphuric acid. It is altogether probable that many if not all soils constantly receive small amounts of sulphuric acid, and it is possible that in some cases the solvent action of this acid on the mineral constituents of the soil may become important.'| Among these substances is chlorine the amount of which in the air varies with the distance from the sea and is greatest at the seashore. On the island of Barbados 116 pounds of chlorine are contributed annually to each acre.” Two or three extractions of soils, however, seldom show the presence of any free acid other than carbonic acid. Carbon dioxide, which is the basis of carbonic acid, is contained in all natural water and in rainfall so that all percolating waters are real acid solvents and exercise a far-reaching effect, a fact now universally recog- nized. That dissolved carbon dioxide may act directly as an acid, thus increasing the solvent power of the water in which it is contained, is probable.? The destructive action of water charged with carbonic acid is most strikingly exhibited in limestone but it is not confined to this type of rock; even quartzose rocks of the ordinary kinds are attacked by it and granite and related rocks are rather quickly affected. Its effect both in the soil and in the zone of weathering ‘ generally is due largely to a reduc- tion of the mass of the hydrates of the hydrolyzed bases by the formation of bicarbonates. ‘The result of its action upon the feldspars is the forma- 1 C. R. Van Hise, A Treatise on Metamorphism, Mon. U.S. Geol. Surv., vol. 47, 1904, p. 205 et al. 2 Harrison and Williams, Jour. Am. Chem. Soc., vol. 19, 1807, p. I. 3 Carbon dioxide is soluble in water to the extent of equal volumes at ordinary temperature and barometric pressure. « The zone of weathering extends from the surface to the ground water (Van Hise, A Treatise on Metamorphism, Mon. U. S. Geol. Surv., vol. 47, 1904). 12 FOREST PHYSIOGRAPHY tion of clay, a most essential element of soils from the physical stand- point, and the freeing of potash, one of the most essential plant foods. In the case of granite rocks the silica set free by the carbonic acid remains partially or wholly in the resulting soils; in the case of limestones the lime at first remains in the form of a carbonate, but potash and soda compounds, which are readily soluble in water, are largely carried away either by percolating water or absorbed by plants. The action of carbon dioxide is also manifest in the formation of carbonates of iron and magnesium. In certain experiments carried on in the laboratory of the U.S. Bureau of Soils some powdered minerals, among which were muscovite and albite, were kept in contact for fourteen months with water and certain solutions in paraffin cylinders. Excepting the results obtained with albite, those obtained by treatment with water saturated with carbon dioxide are so much greater than the corresponding results obtained with pure water that no reasonable doubt can exist that the effect of the carbon dioxide is not only to hasten the rate of solution, but actually to increase the absolute solvent action of the water.! In nature all the elements in the rock and the soil minerals in the zone of weathering are being dissolved all the time, but at variable rates depending (1) upon the strength and abundance of the active compounds in solution and (2) upon the solubility of the constituent minerals. WATER AS A CARRIER In addition to being the substance necessary for the chemical decom- position of nearly all kinds of rocks and soil, water has an important influence in removing large amounts of soluble plant food from the soil. Nearly five billions of tons of mineral matter are annually carried away in solution from the land into the sea, while the amount of sediment is many times greater.” The amount of nitric acid found in drain water (water that runs off through drains, i.e., tiles, etc.) sometimes shows a heavy depletion of the land by the leaching out of this highly important substance. In all drain water lime is found to be leached out most abundantly, mainly in the form of bicarbonate. Magnesia is next in order, then soda and other substances of minor value. Potash is present in drain water in small amounts. Carbonic acid is the most abundant of the acids found in such water, and chlorine and silicic acids are next in order. MECHANICAL ACTION OF WATER AND ICE An important mechanical effect of water is exhibited during rain storms when the erosion of soil on all slopes and its rapid erosion on unprotected steep slopes occur and may lay bare the rock surface 1 Cameron and Bell, The Mineral Constituents of the Soil Solution, Bull. U. S. Bur. Soils No. 30, 1905. 2 E. W. Hilgard, Soils, 1906, p. 24. ORIGIN AND DIVERSITY OF SOILS 13 and enable other soil-making forces again to act upon the exposed rock. Falling raindrops also beat upon and jostle the soil grains or move them about upon the rock surface in such manner as to break off smaller particles, an action which on flat surfaces may tend to increase the amount of soil. It has been computed with reasonable accuracy that 783 million tons of earth and rock measured as soil are removed each year by erosion from the surface of the United States. The amount removed from different watersheds varies greatly, not only on account of differences in the sizes of the drainage areas but also on account of differences in the depth and porosity of the soil, the extent and nature of the vegetable cover, the lengths and declivities of the slope, the rainfall, the temper- ature, the extent of lakes, etc. In the north Atlantic basin the rate is 130 tons per square mile per year; the rate in the Hudson Bay basin is 28 tons and is the lowest on the continent; the southern Pacific basin heads the list with 177 tons. Individually the Colorado River brings down the greatest amount of suspended matter; it delivers 387 tons per year for every square mile of its drainage basin. Practically no suspended matter is transported by the St. Lawrence River, since the water is cleared by sedimentation in the Great Lakes. In general the northern streams carry much less suspended matter than the southern streams, a result due probably to the large number of lakes in the drainage basins of the northern streams, the large extent of bare rock outcrop, and the hindrances to erosion imposed by soil frozen during a large part of the year,! The action of freezing water is due mainly to the expansive force it manifests as it passes from water to ice, and has been described as equiva- lent to the pressure of a column of ice a mile high, or about 150 tons to the square foot. Ifa given rock contains much water in its pore space and is repeatedly subjected to freezing temperatures, the rock will in time be disrupted by heavy internal strains. The extent of the strain effect depends (1) upon the climate, (2) upon the weather conditions, whether uniform or variable, and (3) upon the amount of water con- tained by the various kinds of rock, which in turn differs with the nature of the minerals and their state of aggregation. All rocks when freshly exposed hold by capillary attraction a certain amount of water, which occurs largely as interstitial water. The amounts that may be contained are expressed roughly as follows: granite, 0.37% by weight; chalk, 20%; ordinary compact limestone, 0.5% to 5%; and sandstones from 10% to 12%; while clay may contain nearly one-fourth its weight 1 These computations show that the surface of the United States is being removed at the average rate of .oor3 of an inch per year, or 1 inch in 760 years. Dole and Stabler, Water- Supply Paper U. S. Geol. Surv. No. 234, 1910 (Denudation), pp. 82-83. 14 FOREST PHYSIOGRAPHY in water. The amount in white chalk is as much as 19% and a piece of such chalk may be shattered into fragments by a single night’s frost. The freezing of absorbed water is one of the most general sources of disintegration of building stones. In addition to the expansive force of interstitial water when frozen is the action of freezing water in the joints of the rock, which tends to disrupt large masses from the faces of cliffs and other bare rock surfaces. The effect is heightened if freezing and thawing alternate in periods of short duration. Alternate freez- ings and thawings may be beneficial to the soil after formation, because the freezing of the water in the pore spaces increases the bulk of the (a) whole mass of frozen soil, — an in- crease which is not immediately compensated on thawing, so that aeration and root penetration derive a certain benefit from the process. TEMPERATURE EFFECTS (b) While the agencies we have so far enumerated are active in nature in breaking down rock, soil would be formed without such agencies, though at a slower rate, through the in- herent instability of the rock under changing temperatures. The break- ing of a hot glass plunged suddenly (ec) into cold water is a manifestation of the same force that in humid regions Fig. 1.—Effect of unequal heating of the surface to a lesser extent, in arid regions to sLarek, (o shows theandion ofa kg greater extent, tends to disrupt which the upper portion expands when heated rock masses and to form soil. The arts gets tompentun (Os 2m temperature effects upon Tock gre the average temperature. When contraction Of several kinds: (a) the breaking aed smi a suscety este |apart of large rocks data saul Hise, U. S. Geol. Surv.) masses through expansion and con- traction of the whole mass at an un- equal rate; () the peeling off of rock layers and chips from a rock surface or from the surfaces of bowlders through unequal expansion and contrac- 1 A. D. Hall, The Soil, 1907, p. x1. ORIGIN AND DIVERSITY OF SOILS I5 tion between the surface layer and the layer immediately beneath, a pro- cess known as exfoliation; and (c) temperature changes which expand different minerals at different rates and cause an internal strain to which the rock may finally yield. The first process (a) is so familiar as to require no description. The second process (6) may be understood by the recognition of the high temperatures which bare rock surfaces attain when exposed to the summer sun. On a hot day they may attain a temperature of 150° or 160°, a temperature, so high that the hand can not be held on exposed surfaces without being burned. Between the highly heated surface particles and the particles some distance beneath the surface there must be a zone of shear, for at these high temperatures the surface rock will expand greatly, while the cool rock only a short distance beneath the surface is so much lower in temperature as to occupy smaller bulk, Fig. 1. From observations made near Edin- burgh, Scotland, during 1841-42, the range of earth temperatures at varying depths in soil, sandstone, and trap rock was determined to be as follows: VARIATION OF TEMPERATURE WITH DEPTH} Trap Rock | Sand of Garden | Craigleith Sandstone Depth | : Max. Min. Range | Max. Min. Range Max. Min. Range BIT ee Eat a eee | 52.857 38.88° 13.97" | BA SOw | 937085 a leo Se NS Sanson liso 2524 |rAeOOn Gifeetes-cse ede O7enl4On7.O: 3, 1O120) 1252505. 12 30a55, is af Ouro ls00 lS onO5. | L205 TOBE EME RIAN ves 49.00 | 44.20 4.80 | 50.40 | 43.50 6.90 | 50.30 | 41.60 8.70 QAteetie ene oe). 47.50 | 46.12 1.38 | 48.10 | 46.10 2.00 | 48.25 | 44.35 3.90 Of course the surface inch or two or three inches show much greater ranges; and between the first and twelfth inches the differences may be extreme on hot summer days. The author has noted as the result of temperature observations on loose dry soil during several summer months a maximum difference of 35° to 50° between the first and fif- teenth inches, which was reduced to 5° or 10° before the following sunrise. Translating differences of temperature into units of expansion we have the rate of horizontal expansion varying from .000004825 inch per foot for each degree Fahrenheit in granite to .co0009532 inch in sandstone.’ A change of temperature of 150° in a sheet of granite roo feet in diameter would thus produce a lateral expansion of about 1 inch, an expansion that tends to lessen the cohesion of the rock and to cause a shearing of 1 Trans. Royal Society of Edinburgh, vol. 16, 1849. From G. P. Merrill, Rocks, Rock- weathering and Soils, 1897, p. 184. 2 W. H. Bartlett, Experiments on the Expansion and Contraction of Building Stones, etc. Amer. Jour. Sci., vol. 22, 1832, p. 136. 3 16 FOREST PHYSIOGRAPHY the upper over the lower layers. Although these movements seem slight they are sufficient to produce in time a weakening and breaking of the rock, thus affording a better opportunity for the action of other physical and chemical agencies such as freezing water, bacteria, plant roots, etc.! This form of rock disintegration is most pronounced in massive coarse- grained rocks located in regions of great extremes of daily temperature such as occur in the arid and semiarid portions of the West. When it occurs in homogeneous massive rock it may produce rounded forms or bosses of very characteristic appearance. Stone Mountain, Georgia, 650 feet high, 2 miles long, and 13 miles wide, owes its elliptical dome-like form to such exfoliation along preéxisting lines of weakness in the form of joints. The surface sheets are buckled up in very characteristic forms. They are rarely more than 10 inches thick, but are 10 or 20 feet in diameter. In a few instances small avalanches have been caused by the giving way of such sheets.? The third process (c) of rock disintegration through temperature change, that of crystal crowding, may be understood from the fact that in all rock composed of crystals there is an internal strain due to the unequal rates at which the component minerals expand upon increase of temper- ature. Such expansion has two forms of inequality: (a) the cubical expansion varies with the kind of mineral and (b) the rates along the various crystallographic axes of the constituent minerals are unequal.’ Tt is self-evident that a coarsely crystalline rock under these conditions will disintegrate more rapidly than a rock of finer grain. Rocks of granular structure, all other things being equal, undergo disintegration much more quickly than those in which the individual minerals are closely compacted, as a diabase or a quartzite. It is believed that dark-colored basic rocks tend to respond to the forces manifested by changes of temperature somewhat more readily than do light-colored acidic rocks, because of the more rapid absorption of heat by dark-colored objects in sunlight. It has also been shown that the thermal conductivity of rocks varies in direction according to the structure, being greatest in the direction of the schistosity, where this feature is developed. The result is that in massive, homogeneous rocks the conductivity is the same in all directions, while in finely fissile rocks it may be four times as great in the direc- tion of the fissility as at right angles thereto.‘ WIND AS AN AGENT IN SOIL FORMATION While the action of wind is most clearly seen on the surfaces of land waste where loose dry material may be shifted about in the form of dunes and sand drifts of variable size and shape, wind may also be an important agent in the actual production of soil. Loose particles of rock may be driven through the air against projecting rock ledges, and not only do they themselves tend to become finer through attrition in 1 G. P. Merrill, Rocks, Rock-weathering and Soils, 1897, p. 181. 2 Idem, pp. 245-246. 3G. P. Merrill, Stones for Building and Decoration, p. 410. * G. P. Merrill, Rocks, Rock-weathering and Soils, 1897, p. 184. ORIGIN AND DIVERSITY OF SOILS D7 the air, but they also abrade obstructing ledges. This action takes place on a considerable scale in dry regions and may become one of the most important agents in the denudation of desert lands! It has been estimated that the dust in a cubic mile of lower air during a dry storm weighs not less than 225 tons, while the amount of dust in the same volume of air during a severe storm may reach 126,000 tons.”._‘The great importance of the wind as a soil builder is shown by the wide distribu- tion of the loess deposits of the world. “It would perhaps be an ex- aggeration to say that every square mile of land surface contains particles of dust brought to it by the wind from every other square mile, but such a statement would probably involve much less exagger- ation than might at first be supposed.”?? Dust transportation is not confined to desert regions, but takes place almost everywhere on some scale. 2 Not the least important of the effects of wind on soil has been the wide distribution of volcanic dust as in Oregon, southern Idaho, etc. In the Tertiary period volcanic eruptions took place on a vast scale in many portions of the West. Great quantities of volcanic dust were raised aloft and then deposited at varying distances from the volcanic vents. In many instances such bodies of dust were swept by the wind hundreds of miles from their points of origin, and finally deposited as a sheet of loose waste. Since their original deposition the wind has played upon the surface layers, shifting them about, reworking and redepositing the material, and by these means modifying the qualities of the surface soil of many great tracts. BACTERIA Certain bacteria are able to draw their nourishment directly from the air in a purely mineral environment such as that found upon the surfaces of bare rock, so that even the denuded rocks of high mountains may be populated by these minute organisms. The ragged rocks of high altitudes and steep slopes in the Alps, the Pyrenees, the Vosges, etc., are composed of minerals of the most varied nature, all of which have been found to be coated with a “nitrifying ferment.’’ These bacteria develop by absorbing carbonate of ammonia and vapors of alcohol from the air and they even assimilate carbon dioxide. The wide distribu- tion of these organisms, their great number, and the manner in which 1 W. Cross, Wind Erosion in the Plateau Country, Bull. Geol. Soc. Am., vol. 19, 1908, pp. 53-62; S. Passarge, Die Kalahari, 1905; W. M. Davis, The Geographical Cycle in an Arid Climate, Jour. Geol., vol. 13, 1905, pp. 381-407. 2 J. A. Udden, A Geological Romance, Pop. Sci. Mo., vol. 44, 1898, pp. 222-229. 8 Chamberlin and Salisbury, Geology, vol. 1, 1904, p. 22. 18 FOREST PHYSIOGRAPHY Fig. 2.— Underground burrows and chambers of earthworm population, in well-drained sand bank near New Haven. The vertical burrows connect with underground chambers showing as dark lenses within an inch and an inch and a half of the bottom of the photograph. Seven or eight can easily be identified. At the time the photograph was taken many of the chambers and some of the burrows were completely filled with dark humus collected by the earthworms. When the humus was scraped away it was seen that each chamber had a smooth and generally flat floor and an irregularly arched roof. Such of the openings as had been filled with humus had been abandoned. The open chambers and burrows were teeming with earthworms and with insects of many genera. The upper edge of the bank was broken down during a dry spell and at a time when the ground was fully stocked. The scale in the upper right hand corner is three inches long. (Photograph by Mason.) ORIGIN AND DIVERSITY OF SOILS 19 they prepare the rock for the microscopic vegetation which is usually found in the form of a layer covering rock surfaces and soil par- ticles, place them among the important geologic agencies that have effected the disintegration of the rock and the formation of soils.! Their action is carried on on a small scale in cold temperatures and on a very large scale under normal temperatures where the rock is but thinly covered with earth. The action is also carried on among rock fragments, tending gradually to reduce them. The bacterial organisms penetrate every cleft or crevice, and by the chemical action of their secretions continually reduce the rock to smaller and smaller sizes, acting even on the most minute fragments. Each rock particle is found covered with a film of organic matter accumulated by these bacteria and the plants of a higher order which secure through them a food supply. ANIMALS AND HIGHER PLANTS AS SOIL MAKERS The most active animal agencies in producing and modifying soils are earthworms, which influence the physical state of soils by making them more porous and open. Darwin’s studies showed that the intesti- nal content of worms has an acid reaction which has an effect on the soils passing through the alimentary canal. They still further modify soils by drawing into their holes leaves and other organic materials which gradually decay and are converted into humus, Fig. 2. Darwin estimated that an average of about 11 tons of organic matter is in this manner annually added to each acre of soil in regions where earth- worms abound. Earthworms tend also to increase the amount of ammonia in the soil, to make the soil finer by attrition during the process of digestion, and to mix the soil by bringing to the surface portions of the subsoil. They thus play a notable part in maintaining soil fer- tility. The excreta of earthworms contains more nitrogen and other easily oxidized compounds than the original soil, and after excretion by worms soil contains phosphoric acid and calcium carbonate in more readily soluble forms.” It has been observed that when earthworms are drowned out the surface soil layer remains compacted and vege- tation grows very feebly until earthworm immigration has restocked the soil. The action of burrowing animals, worms, insects, etc., in the soil, allows freer access of rain water and drainage water and increases the depth and the rate of rock and soil decay. Ants sometimes produce 1H. W. Wiley, Principles and Practice of Agri. Analysis: Soils, vol. 1, 1906, pp. 39-40. 2 Cameron and Bell, Mineral Constituents of the Soil Solution, Bull. U. S. Bureau of Soils No. 30, 1905, p. 41. 3 A.D. Hall, The Soil, 1904, p. 150. 20 FOREST PHYSIOGRAPHY highly important effects upon the soil, especially in tropical countries as in Brazil, where they often build tunnels hundreds of yards long which give the soil-making forces access to the subsoil. They carry great quantities of leaves into their nests and by this means and by their excreta contribute vegetable acids to the soil and thus promote rock and soil decay.' More general in their action upon both rock and soil are plants and plant roots. Roots force themselves into the crevices of rocks and minerals, they wedge apart rock masses, and thus expose new surfaces and a larger extent of surface to the action of other forces. "The mechan- ical force exerted by growing roots is very great. The root of the garden pea has a wedging force equal to 200 or 300 pounds per square inch, a force that is exerted without harmful effects upon the small roots because of the protection afforded by the corky layer of the root tips.? Vege- tation also assists in rock decay by shading the surface and permitting the retention of a larger amount of water upon the immediate surface where it exercises a dissolving action as already described. The organic matter contributed to the soil by decaying vegetation promotes rock decay by furnishing carbon dioxide or carbonic acid. Rootlets of plants in contact with limestone dissolve large portions of this rock through the solvent action of root moisture. The action of rock disintegration is also extensively carried on by the lower forms of vegetation, among which the lichens produce the most important-effects. The rock sur- face is corroded and a soil cover originated. A prominent ingredient of the lichens is oxalic acid, an acid that compares in strength with hydrochloric and nitric acids, and that powerfully aids rock decay. Plant roots that permeate the soil are agencies of oxidation which have a very appreciable effect in altering some soil constituents and influencing soil fertility. Active extracellular oxidation is carried on by the plant roots chiefly by means of the enzymes (the name applied to any unorganized chemical ferment such as diastase, pepsin, etc.) which they excrete, and not to organic (carbonic acid alone excepted) and inorganic acids which they were formerly supposed to excrete. It is believed that the effect of the roots of growing plants in dissolving the organic substances of the soil is due chiefly to this action. Among phanerogams extracellular oxidation is strongly localized and limited to the absorbing surface of the root; the most intense oxidation is effected by the root hairs. Oxidation is more marked when an optimum water 1 J. C. Branner, Ants as Geological Agents in the Tropics, Bull. Geol. Soc. Am., vol. 7, 1896, p. 255; idem, Geologic Work of Ants in Tropical America, Bull. Geol. Soc. Am., vol. 21, IQ10, Pp. 440-406. 2 E. W. Hilgard, Soils, 1906, p. ro. ORIGIN AND DIVERSITY OF SOILS 21 content is maintained in the soil, while saturation of the soil produces a decided depression in the rate of oxidation. Oxidation by plant roots is increased by the presence of calcium salts, potassium salts, phos- phates, nitrates, etc. The action of plants in promoting the formation of soil is well illus- trated by the manner in which plant associations succeed each other upon extensive areas of bare rock, a succession that is dependent upon the ability of each plant group to live under the hard conditions which exclude the next higher group. Crustose lichens are the only plants which are able to establish themselves on a bare rock face. Upon the thin soil formed by these, other lichens are able to secure a foot- hold. Then appear the mosses, for example Hedwigia, Grimmia, etc., which eventually eliminate all but the erect (fructicose) lichens. The mosses still further increase the soil layer, both through the accumu- lation of mineral particles and by their own decay, and are in turn wholly or partly displaced by the more xerophytic species of ferns such as the spleenworts. With the ferns appear many herbaceous flowering plants, notably the stonecrops and saxifrages, and certain grasses. In addition to the carbonic acid which is excreted by the roots of all plants many of them secrete vegetable acids such as oxalic and citric acids which assist in soil formation. The herbaceous plants in turn prepare the way for, and are eventually succeeded by, shrubs and trees. Senft describes the vegetation that takes hold of landslips and coarse terrace deposits (near Eisenach) and shows how it undergoes great changes in type due to soil changes brought about chiefly by the vegeta- tion itself. In the locality examined the bare stony heaps were first clothed with mosses, then xerophytic grasses; later other xerophytic herbs came in and also shrubs like the juniper which gave rise to dense, bushy growths. In twelve years an impenetrable bush land had arisen and finally Sorbus, Fagus, and other trees appeared and a forest arose. During this change in vegetation type the soil had constantly changed and improved. Each kind of vegetation suppressed another — bush land vanquished xerophytic grasses, and forest vanquished bush land.” The effects of minor rock structures upon plants is extremely inter- esting. Peculiar and variable rock habitats induce variations in plant societies which are due chiefly to local differences in the nature of crev- ices — joints, fissures, etc. — in the rock. 1 Schreiner and Reed, The Role of Oxidation in Soil Fertility, Bull. U. S. Bureau of Soils No. 56, 1907, pp. 7-9 et al. 2 Quoted by Warming, (Ecology of Plants, t909, Ox. ed., pp. 352-353. 22 FOREST PHYSIOGRAPHY ““Some of these receive water percolating from higher parts of the mountain, and may remain moist throughout prolonged periods of drought; other crevices obtain their water exclusively from the strictly local rain. Some crevices contain abundant detritus and are therefore endowed with a greater power of storing water; others are poor in detritus and allow the water to pass away. The chemical composition of the detritus also varies, as some crevices contain abundant humus, in which numerous earthworms may lurk, whereas others are poor in humus. Cracks in rocks supply an endless variety of habitats, each of which forms a special kind of environ- ment.’ 1 The point is of exceptional importance in regions of thin soil where the lower roots of all plants and all the roots of some plants are in inti- mate association with the rock. ‘Talus slopes frequently show similar variations. Their lower slopes or lower margins are commonly wooded because of the finer rock waste and the greater amount of water which here reappears. The loose upper slopes are treeless except where the talus has become temporarily stable and clogged with finer waste or where a change of geologic formation or the arrangement of crevices cause seepage, a common condition at the upper edge of talus slopes. THE CAUSES OF SoIL DIVERSITY The preceding discussion will enable us to see that the soil is an ex- tremely complex mixture. There are present mineral débris from rock degradation and decomposition; organic matter, the partly decom- posed remnants of former plant and animal tissues; the soil atmosphere, always richer in carbon dioxide and generally in water vapor than the atmosphere above the soil; living organisms, such as various kinds of bacteria, and often molds, ferments, and enzymes; and finally the soil water, a solution of products yielded by other substances. This enumeration is sufficient to show how diverse are the origins of the different components, and to suggest how varied are the reactions that take place in the soil even after its formation. These facts need emphasis because of the general view that soil is mere dirt or rock waste, or that it is everywhere the same, whereas the truth lies nearer the other extreme. The soil is a great complex of varied elements, formed in many ways, and subject to the most widely diverse changes after its formation. Nor have we exhausted the list of diversifying forces. We have yet to consider briefly the transportation of rock waste after formation, the various agencies concerned in it, and the results upon the soil texture and fertility. All soils are subject to some movement at the earth’s surface, and since the soil particles are of different sizes, weights, and compositions, they must respond in different ways to the forces that tend to move 1 E. Warming, Ecology of Plants, Ox. ed., t909, p. 245 ORIGIN AND DIVERSITY OF SOILS 23 them. The simplest case is that of deposition by running water, with gradually diminishing velocity, where on the whole the coarser particles are deposited first and the finer last. The action and the result are so familiar to all as to require no extended discussion. The distribution of material in an alluvial fan or a delta or a flood plain always follows this well-recognized law. The effects of creep and rainwash are not so simple. Under the influence of constantly changing temperature all hillside or colluvial soils tend to move down slope. Contributing toward the same result is the action of percolating soil water and ground water, cultivation, etc. In all these cases the fundamental and ultimate force is gravity, but because gravity is manifested in so large a number of forms it is clearer to consider the forms themselves and not the basic force on which they depend. In such cases of creep there is not that suspension of particles in water that permits thorough stratification or sorting, consequently coarse and fine are left mixed together, and the rate of movement may be so slow as scarcely to be perceptible. Where rock formations succeed each other in short distances soil creep may cause an important lack of sympathy between the underlying rock and the overlying soil. The soils of the higher may come to rest over the lower formations, and the rock character give little clue to the nature of the overlying soil. Cases of this kind are frequent in the Appalachian Mountains and the ridges of the Great Appalachian Valley, and are especially well marked where the boundary between two unlike forma- tions occurs on a hill slope. If the slopes are quite steep and the rock formations numerous on a given slope, such overplacement of soils may produce extreme effects and the waste from the different rocks become “so mixed as to show at the foot of a slope or on the inner border of a foreland plain but little relation to any particular rock. On broad plateaus where the boundaries between rock formations are far apart mixtures of soil types take place to an important degree only near the boundaries of the formations; over the greater part of the outcrop of a given formation there is a close relation between the underlying rock and the soil. The Colorado Plateaus of the Southwest and portions of the Cumberland Plateau furnish excellent examples of this law. The rate of movement has been made the basis of a classification of soils according to origin that deserves a word of explanation. While all soils are subject to some movement, the movement may be so slow as to be of no importance, as on portions of flat tablelands or base-leveled areas like the Piedmont Plateau of Georgia and Maryland. Such cases of no movement or of little movement will tend to cause a certain sympa- thy between soil and rock in a given locality, and the minerals that occur 24 FOREST PHYSIOGRAPHY in the rock are found in the soil or at least their decomposition prod- ucts are found there. Such soils are sedentary or residual soils, and the term generally implies a fundamental relation between a rock terrane and the soil covering it. On the other hand, if the soil has once been in the grip of a transporting agent such as the wind, running water in the form of river or brook, or a continental ice sheet, or a glacier, it is considered a transported soil; and the term éransported always connotes a mixture of soil ingredients in the case of ice, and sorting in the case of water and wind. By these agencies soils may come to rest far from their place of origin. The alluvial deposits of the Mississippi flood plain are derived from fully one-fourth the whole United States. The underlying rock now perhaps deeply buried in a given locality may be limestone, but the alluvial soil overlying it may be deficient in lime and in still other ways less fundamental bear little or no relation to it because it was not derived from it to any important degree, or perhaps to any degree at all. No less true are these statements when applied to: wind- borne soils. In every dry region whirlwinds raise aloft great clouds of dust that settle down near by or become more or less permanently lodged perhaps hundreds of miles away in extra-desert regions, where their relation to the rock or soil they overlie is purely fortuitous. The great loess deposits of western China, the dust soils of Oregon and Idaho, the loess deposits of the Mississippi Valley, all are illustrations of wind- borne soils, though they are not all fundamentally related to the extremes of arid conditions. Glacially transported material has or may have the same discordance with respect to the underlying rock. Over the sandstone and limestone areas of the Great Lake region has been swept glacial detritus in vast amounts, and although the material reflects the character of the under- lying rock to a notable degree (perhaps on the average about 80% of it is locally derived), yet an important share is also derived from northern localities. Bowlders of granite, gneiss, greenstone, slate, and basalt may be found scores and even hundreds of miles from their nearest out- crop, and everywhere in the glacial till are found important amounts of clay which were derived at least in small part from northern localities. The effects of the continental ice cap on the soil are discussed in greater detail in succeeding pages, and need not be further described here. Alpine glaciers have had less important effects upon soil because of their relatively slight development and because they have produced their effects chiefly in mountain valleys where their deposits are so restricted that though they may be important to the farmer they are relatively of less importance to the forester. ORIGIN AND DIVERSITY OF SOILS 25 The part that the various agencies concerned in soil formation have played in the making of the soil of North America is brought out in the following table.+ DISTRIBUTION OF SOIL TYPES BY REGIONS ® (The surface of each continent is taken as 100) aS Salle 3 a glo 48 ZH\OE| a [ 1 E. W. Hilgard, Soils, 1907, p. 346. 2 For a discussion of lime effects see E. W. Hilgard, Soils, 1907, pp. 378-381 et. al. CHEMICAL FEATURES OF SOILS 69 deteriorate, in contrast to the Corsican pine, which is a lime-loving tree. The higher the clay percentage of a soil the more lime carbonate it must contain in order to exhibit the advantages of a calcareous soil. In sandy lands a characteristic lime growth may reflect the presence of only 0.10% of lime, while in heavy clay soils 0.6% is required to produce the same result. This explains the prominent color of dark-tinted humus in sand soils when very small amounts such as 0.2% of lime occur, whereas a comparable effect is produced in clay soils only when the percentage rises to 1%. A soil that effervesces with acids contains at least 5% of carbonate of lime, and percentages so small as to make the soil distinctly calcareous are not distinguishable by this mode of analysis. In the study of the effect of lime upon vegetation a difference of opinion has arisen, probably due to the very different methods pursued and the different regions in which the students have worked. Hilgard, who has done the most extensive work of this character in America, concludes that the moisture of the soil is the point of first Importance in the dis- tribution and welfare of vegetation but that the condition next in impor- tance is the amount of lime present. He grants, of course, that certain species are indifferent to lime, but holds that most species respond to lime in a marked manner and that on the whole the presence of lime tends to greater fertility except in the obvious case where it is present in excess. He finds that in Mississippi and Alabama there is a marked correspondence between the growth habit and types of trees and the geologic formations, so that a geologic map of the region would also be to a large extent a map of the various tree zones. The conclusions of Hilgard are of great value because they were formulated as early as 1860 after extensive study of native vegetation which grew in a soil that was almost undisturbed by man, hence a vegetation that represented a long term of adjustment to the soil. This is obviously an advantage over the conditions in Europe, where the observers of the vegetation have quite constantly to eliminate the influence of cultivation and other disturbing influences due to the long occupation of the land by man. It would not be fair, however, to dispose of the matter by merely stating the ground of Hilgard’s conclusions. The contentions of the European students are cogent and interesting and deserve equal attention here. Warming says: “Although the characteristics of the lime-flora are clear and distinct, yet in the past the influence of lime upon vegetation has been overestimated. Indeed, a distinction has been made between calciphilous and calciphobous plants. Recently it has been definitely established that the amount of lime in itself, in so far as it does not operate physically, can not be the cause of differences in the flora, for not only can calcicolous plants be cultivated in soil that is poor 7° FOREST PHYSIOGRAPHY in lime, but silicicolous plants, and even bog-mosses, which are regarded as preeminently calciphobous, can grow vigorously in pure limewater if the aqueous solution be otherwise poor in dissolved salts. It has been overlooked that nearly all lime soils are rich in soluble mineral substances, and this wealth excludes plants belonging to poorer soils; beyond this the important physical characters of calcareous soil, compared with granite soil, come into play.”’! In general the disagreement of the conclusions as to the power of lime to control plant distribution appears to be due to the absence of a standard conception as to what constitutes a lime soil. By some a soil is regarded limey only when it effervesces with acids, yet not less than 4% of calcium carbonate in a soil will respond to the acid test, whereas 0.1% will have important effects 6n plants provided the soil is sandy. So far, most of the conclusions have been mere obiter dicta. The conclusions of American investigators, based on native and practically undisturbed vegetation, are essentially sound, though they require important modifications where the physical conditions become extreme. MAGNESIUM Magnesium forms an essential part of the rock known as dolomite and may exist as a silicate in such rocks as serpentine, talc, etc. In igneous rocks it occurs in the minerals pyroxene, mica, olivine, etc. Magnesia is invariably and rather abundantly found in the seeds of plants and is a very important plant-food ingredient, but must not occur in excess or it will cause, through chemical action, a pronounced change in the capacity for imbibition, and thus particularly disturb the functions of the plant. Magnesia is especially concerned in the transfer of phosphoric acid through the plant tissues; while magnesia predomi- nates in the fruit of a number of crop plants lime predominates in the leaves, so that there is apparently a connection between the extension of leaf surface and the lime requirement. SODIUM Sodium occurs in largest percentage in the igneous rocks as a con- stituent of the soda-lime feldspars, amounting on the average to 214% of the igneous rocks. These feldspars are more readily attacked by water and carbon dioxide than are the other common minerals save certain basic silicates, so that the whitening and the softening of felds- par is one of the first signs of rock decay. The sodium salts are so soluble, however, that they are leached away almost as rapidly as formed, with the result that soils are normally rick in potash but poor in soda. In poorly drained soils of arid lands, however, these qualities result in * E. Warming, (Ecology of Plants, Ox. ed., 1900, p. 58. CHEMICAL FEATURES OF SOILS 7a: a concentration of soda through the continued evaporation of ground water, where as carbonate, sulphate, and chloride it gives rise to alkali tracts poisonous to all but specially adapted species of plants. In northern Chile, one of the most arid tracts of the world, the sodium has become concentrated in nitrate deposits. These on account of the fixed nitrogen which they contain are extensively mined and shipped to many agricultural regions in humid lands as a fertilizer. Thus through its chemical properties — the extreme basicity of the element and the solubility of its salts— sodium causes the most arid deserts to add to the fertility of the garden spots of the world. Sedimentary rocks, the accumulations of the constituents of soils of former ages, are usually deficient in sodium and may be almost free from that element. Their soils, the result of a second cycle of leaching, tend to be still more barren in sodium. The use by land plants of potassium is doubtless an adjustment to the prevailing composition of soils; marine plants, living in an environment where sodium is dominant, show a parallel use of sodium in their tissues though they use potassium also, to some extent. POTASSIUM One of the most important elements of the earth’s surface from the standpoint of plant growth is potassium, which in the form of a nitrate is found in nature (saltpeter). It occurs in small and large quantities in a great variety of igneous and metamorphic rocks, but may be absent from sedimentary rocks. It is present in mica, amphibole, and pyroxene, and when combined with silica is an important member of orthoclase and other minerals. Granite soils generally contain a good supply of potash on account of the common occurrence of potash feldspar in them. Granite soils may be deficient in lime, however, unless hornblende is present, since lime-feldspar is not likely to occur as an accessory ingredient of granite. The amount of potash necessary for high soil productivity is about 0.5% and at this figure the addition of potash has but little effect upon the fertility. At 0.25% there is a deficiency that must be made up by fertilization. These figures do not apply, however, in arid or tropic lands. In tropic lands the prevailingly high temperature, the great rainfall, and the continuous leaching, cause a very rapid liberation of potash from its insoluble form as well as its rapid removal; smaller amounts are there- fore necessary at any given moment. In arid regions the absence of rapid leaching allows the accumulation of earthy salts of many kinds, among which potash is prominent. 1 E. W. Hilgard, Soils, 1907, pp. 354-355. he FOREST PHYSIOGRAPHY PHOSPHORUS Of very high importance in soil fertility is phosphorus, found in the minerals vivianite and apatite, in the bones of animals, and in the seeds of plants. Apatite (phosphate of lime) is an almost universal constituent of granitic rocks, but occurs in very small quantities! The amount of phosphoric acid (P2O;) contained in granitic rocks rarely exceeds 0.2% and may fall as low as 0.05%; but small as the amount is it probably is the main source of supply of phosphates existing in the soil. Phos- phorus is most abundant in the basic eruptive rocks such as diorites and gabbros and deficient in such rocks as sandstone and slate. Where the minerals vivianite and apatite are abundant in the country rock, as in the basaltic lavas of Hawaii, phosphoric acid may be pres- ent in the soil in exceptional amounts, — nearly 2%. Unfortunately in this particular case it occurs in the form of an insoluble basic iron compound, ferric phosphate, which is dissolved with such difficulty that it is wholly unavailable to vegetation and the soil containing it is actually phosphate poor. The same is probably true of certain ferruginous soils in California and the South. The lower limit of adequacy of phosphoric acid in the soil is about .05%. Exceptionally soils may contain as much: as .30%, while .15% is regarded as adequate. In non-ferruginous lands the amount required is smaller than in the case of ferruginous lands because the iron renders phosphoric acid inert by forming ferric phos- phate, an insoluble substance.” Phosphate deposits are derived chiefly from animal remains, but animals derive it from plants, which in turn depend for their supply upon the alteration products of apatite. Commercially important deposits of apatite occur in Spain, Canada, and Norway. From the Norwegian deposits a commercial fertilizer is now manufactured, a phosphoric-chalk manure. Phosphorus is one of the rarest essential plant foods, and its conservation should be a matter of great concern. Sewage contains relatively high percentages of it, and the application of sewage to the land instead of its wastage in rivers and the sea is one of the most important though as yet limited uses of this neglected fertilizer. SULPHUR Sulphur plays an important part in the nourishment of plants, since it is an essential constituent of vegetable albumen and allied compounds, 1 The pure crystalline mineral apatite rarely occurs in large masses. Minute crystals of it are found widely scattered in many rocks, granite, basalts, etc. The largest deposits occur in connection with carbonate. of lime in rocks known as phosphorites which closely resemble limestone. Extensive phosphate deposits are found in southern California, Florida, Alabama, Tennessee, Kentucky, Wyoming, Utah, Idaho, Montana, etc. 2 EK, W. Hilgard, Soils, 1907, pp. 393 et al. CHEMICAL FEATURES OF SOILS 73 hence a soil to be fertile should always contain sulphates in available form. The usual form of occurrence is in combination with the metals to form sulphides or with oxygen and a metal to form sulphates. It is an essential part of the mineral pyrite, and when combined with oxygen and calcium forms the valuable fertilizer gypsum. The amount of available sulphur existing in any soil is usually very small. The relative available amount in ordinary soils is indicated in the following table, in which it is assumed (1) that the mean dry weight of a surface foot of soil is 80 pounds, and (2) that the amount of soluble material is accurately represented by the results of a large number of analyses collected in the Tenth Census Reports. RELATIVE AMOUNTS OF CERTAIN ESSENTIAL PLANT FOODS IN AN ACRE-FOOT 1 IRotasheinksurbace Toot sperMacre cc dia. mw) neccmanyeoee ca dn Ose Sul 2a cin 3.76 tons Sodas mySuLbpacetOOt* PELAACKE Veisks arom wer eee ncaa eters Llane creates cen Me THES Goins Miagiesiainusuriace foot, Wem acreyy aya vit ten ih see uee wales Se aaah eal BOD T os Pimepines ucla ce TOO fy Der saGLe! naa ucciennievele: osama a tuonee | ayia we ne gnMl hay nel 1.88 Phosphoric acid in surface foot, per acre...... EATS eee see O en ECS OS ESE 1.07 Sulphuriciacid: inisurface foot, sper acre (i. Vi is Ss le kins SOL ihe Soluble; silicayin;surface foots peracre 42 2226 ee ie Bil OWN In the decomposition of the various compounds of iron and sulphur, oxi- dation affects either or both the iron and the sulphur; when the iron alone is oxidized the sulphur or some part of it separates as hydrosulphuric, sulphurous, or sulphuric acid. The sulphur of the hydrosulphuric acid may be later oxidized to sulphurous or sulphuric acid through the action of water and oxygen with or without the assistance of bacteria, though bacteria are often the inciting cause of the oxidation. In the form of sulphuric acid sulphur is immediately available to plants, indeed sulphuric acid itself is found in the tissues of plants in small quantities. It is subject to steady depletion in this form by percolating water and by combination with bases in the formation of sulphates, in addition to the demands upon it by growing vegetation.” TotaL PLANT Foop; AVAILABLE PLANT Foop The data concerning plant food are to be distinguished from those derived by mere chemical analyses of soils, which in themselves are of small value in understanding ecologic conditions. Almost all soils show on ultimate chemical analysis an abundance of the elements re- quired for almost any given crop, but there is the widest difference between the forms in which the elements occur, so that it is the available plant food, and not the ultimate amount of plant food that may be produced on complete decomposition, that is a matter of chief interest 1 F. H. King, The Soil, 1905, p. 102. 2 C. R. Van Hise, A Treatise on Metamorphism, Mon. U.S. Geol. Surv., vol. 47, 1904, p. 468. 74 FOREST PHYSIOGRAPHY in the study of the chemistry of soils. The soil must be regarded as possessing most of its plant food in such a state of combination that it can not be utilized by the plant directly, but must by weathering slowly pass into the soluble, 1.e., the available, form. DETERMINATION OF SOIL FERTILITY The approximate chemical nature of an ordinary soil may be ascer- tained in a direct manner by the determination of both the decomposed and undecomposed minerals present in it. The determination of its fundamental nature requires an examination of the undecomposed minerals only, since it is presumed that the decomposed part of the soil has been derived from the constituent minerals and since the undecom- posed material forms by far the greater bulk of the soil. But such an analysis is of less value than direct qualitative and quantitative chemical analyses of the soil character. Even the latter analysis does not always furnish a reliable guide as to the productivity of the soil. The previous history of the land and the physical characters of the soil may be predominating factors. In attempting to ascertain the nature and amount of the decomposed portion of the soil various working pians, which attempt to imitate plant action, have been tried. The water-soluble ingredients of the soil are only a portion of the total number of substances upon which plants may draw for food, because the plant roots act not alone through the me- dium of water but also through water charged with carbonic and possibly other acids. Clearly, then, the action of plant roots may be imitated more closely by employing in the analysis a weak acid solvent that will act upon the soil in a manner similar to the soil acids. The weak acid solvent employed is empirically determined, for no one has yet analyzed the soil about a growing plant in such a way as to ascertain under pre- cisely what conditions the various soil acids act. The results of soil analyses by means of weak acid solvents must be compared with cultural experience and observations on natural plant growth; the results are thus empirical approximations, but they are the best that have been achieved. It has been found by such observations that all soils are continuously soluble to some extent but that the differences between the solutions derived from soils of low and of high productivity are very striking. Plants differ very greatly in the energy and quality of their action upon reserve soil ingredients, so that no single solvent used in an analysis could properly represent the action of plant roots in general. Among the many solvents employed for the purposes of soil extraction and the determination of immediate soil productivity are citric acid (a 1% solu- CHEMICAL FEATURES OF SOILS a5 tion is most commonly employed) and aspartic acid, among the weak acids; and hydrochloric, nitric, and a few others, among strong acids. The hydrochloric acid is employed in densities ranging from 1.1 to 1.16, while a density of 1.115 has been found most convenient and satisfactory of all. From experience with acids of different strength it has been found that a five-day period of digestion with hydrochloric acid (density, 1.115) is sufficiently effective in showing what plant-food ingredients of the soil maintain its permanent productive capacity. This appears to be the natural limit of the action of the acid upon the soil and pro- duces a maximum effect. There is much to justify the contention that the only legitimate solvent in determinations of soil fertility is carbonic acid, the commonest, the natural, and the most abundant acid in the soil.' The immediate soil requirements may also be empirically deter- mined with a fair degree of accuracy by analyzing the ash of the vege- table growth and establishing a ratio between the normal ash ingredient and the actual soil ingredients. It has been found that in the case of a deficiency of certain kinds of plant food there is a disturbance of nutri- tion that manifests itself in the form or in the development of the plant and affords a direct basis for future determinations of a similar sort without the repetition of the full chemical analysis of the ash. Unless extreme physical characters interfere with normal plant growth virgin soils showing high percentages of plant food as determined by extraction with acids? (hydrochloric, nitric, etc.) invariably prove highly productive. Hilgard states the law as follows: ‘‘The actual amounts of soil ingredients . . . rendered accessible to plants are . . . more or less proportional to the totals of acid-soluble plant-food ingredients present.” ® The natural condition which this result indicates may be stated thus: the larger the total amount of plant food in the soil and ultimately available the greater will be the immediate effect of the weathering agen- cies that produce available plant food. This conclusion appears to be rather well established and indicates how high a degree of importance should be attached to the chemical composition of soils despite the relatively higher importance of physical qualities in general. Neither group of qualities can be adequately applied to a soil type to the ex- clusion of the other group. All the field conditions must be evaluated before a soil analysis can be regarded as complete. It is in the highest degree unscientific longer to advocate the supreme importance under all circumstances of a single group of soil qualities. 1 A. D. Hall, The Fertility of the Soil, Science, n. s., vol. 32, 1910, p. 366. 2 E. W. Hilgard, Soils, 1907, pp. 343-353, etc. 3 Tdem, p. 346. 70 FOREST PHYSIOGRAPHY HARMFUL ORGANIC CONSTITUENTS OF THE SOIL In concluding this discussion of some of the chemical properties of soils it seems desirable to note an important recent result in one of the most complicated branches of soil chemistry, the identification of harmful organic substances that occur in the soil, the determina- tion of their properties and the formulation of remedies to offset their effects. It has been found that many plants can possibly excrete, as the result of growth, organic compounds which are poisonous to the plants producing them. It has been found that these organic substances while inhibitory to the plants which produce them have little or no effect upon other plants. It is therefore concluded that they may be positive forces in producing a natural succession or rotation of wild vegetation not explained by any change of soil or climate.! These harmful substances are constantly being added to the soil, and some of them are injurious in quite small amounts and are a cause of infertility even when the amount of plant food in the soil is abundant.’ The most markedly harmful bodies are found within the plant but are not apparently an essential part of the life of the plant. Of this class are arbutin, vanillin, heliotropine, ter- penes, etc., all of which are very injurious. Tyrosine is injurious in quite small amounts, while some of the protein decomposition products are not only harmless but appear to act as plant nutrients; such for instance are asparagine and leucine.s The harmful substances havea toxic effect on the plants, but are destroyed or rendered harmless by other substances like soda and lime in cooperation with plant roots. Oxidation in a high degree converts some of these bodies into harmless substances, but a low rate of oxidation causes the organic matter to decompose incompletely and gives rise to organic compounds unfavorable to plant growth.4 Hall suggests that these so-called toxic substances may be normal products of bacterial action upon organic residues in the soil and that as such they may be as abundant in fertile soils rich in organic matter as in the sterile soils from which they were extracted. He points to the great power of the soil in precipitating soluble materials within it as a possible natural remedy for such toxic substances.° It has also been found that tree roots have a toxic effect on the growth of wheat. The absence of grasses about certain trees is thus attributed not to depletion of plant food alone but to the toxic effects of the tree roots heightened by shade and the well-known injurious effects of washings from trunk and leaves.‘ 1 Schreiner and Shorey, The Isolation of Harmful Organic Substances from Soils, Bull. U.S. Bur. Soils No. 53, 1909, pp. I-3. 2 Tdem, p. 20. 3 Idem, pp. 12-13. 4 Schreiner and Reed, Certain Organic Constituents of Soils in Relation to Soil Fertility, Bull. U. S. Bur. Soils No. 47, 1907, p. 13. 5 A. D. Hall, the Fertility of the Soil, Science, n. s., vol. 32, 1910, p. 367. 6 C. A. Jensen, Some Mutual Effects of Tree Roots and Grasses on Soils, Science, n. s., vol. 25, 1907, pp. 871-874.’ CHAPTER VI HUMUS AND THE NITROGEN SUPPLY OF SOILS SOURCES AND PLANT RELATIONS THE nitrogen of the soil is a matter of paramount importance in fores- try, especially in the maintenance of a proper forest growth and in efforts at reforestation, for as we have seen a supply of nitrogen in some form or other is absolutely indispensable to plants. We shall therefore discuss it somewhat more fully than the other plant foods of the soil. To be available to plants nitrogen must be in soluble form, and it is therefore as a nitrate that it is used by the plant. The main source of nitrogen is humus, whence it is derived chiefly by bacterial action; although nitrogen exists in unhumified organic matter it is not in an available form. Other sources of nitrogen are (a) nitrogen-fixing bacteria that live in a free state in the soil and derive their nitrogen from the soil air, (6) nitrogen- fixing bacteria that live in symbiotic association with legumes and other plants, and (c) rain water. The amount of nitrogen contributed to the land in the last-named manner amounts to a half pound or a pound or more per acre with a rainfall of about 30 inches per annum.’ NITROGEN BROUGHT TO THE SURFACE OF THE EARTH BY RAIN (Pounds per acre per annum.) Nitrogen Locality ix 5 Remarks MODS Nitric Total acal Rothamsted, England................. 2.823 0.917 3.74 In 1888-89 Barbad oSteset etait eh oe ae I.009 2.443 3.452 5 years’ average Britishia Guan ays see ee seas ese I.351 2.190 3.541 7 years’ average [Kea SalS Seek en een ees alae aS 2.63 1.00 3.69 3 years’ average Witala sees areas hero ees i Eons 5.00 - 350 5-42 Do. Nitrogen is usually the first element to become exhausted in the soil because the nitrates are exceedingly soluble and no part of the soil has any special power of holding back nitric acid when it passes in aqueous solution through its pores, so that the nitric acid produced in the soil 1 A.D. Hall, The Fertility of the Soil, Science, n.s., vol. 32, rg10, p. 368. 2 H. W. Wiley, Prin. and Prac, of Agri. Anal.: Soils, vol. 2, 1906, p. 448. UE 78 FOREST PHYSIOGRAPHY passes at once into the vegetation, or remains in store in dry periods, or passes into the drainage water and is lost.' It is important that the forester retain the humus of the forest soil and increase it or make the amounts already there more useful, since the nitrogen which it yields is one of the rarest of the essential plant foods in the soil. The maintenance of humus in the soil requires a forester’s constant attention to renewal of growth after cutting, proper drainage, a shaded surface, etc.? Different species of plants demand very different amounts of humus: some plants appear to require none at all, as those that develop on bare rock; some require a moderate amount, as is the case with certain grains; and others, notably the moorland plants, thrive only in rich humus, and have special methods of nutrition dependent upon the kind of soil in which they live.® The great value of the birch and the aspen lies in their power of rapid dispersion and quick germination and growth in sterile soil or soil robbed of humus by repeated fires. They thus prevent excessive erosion, which is usually so destructive of the soil after fire has destroyed either or both the forest cover and the soil humus. In this manner and by their rapid growth they often afford an opportunity for the seedlings of longer- lived and more valuable trees to come in under conditions that insure their successful growth.# As an illustration of the importance of humus in maintaining an original growth of vegetation may be cited the fact that a change in the forest vegetation is known to have occurred in Denmark during past millennia, owing in part to the action of the wind, which by blowing leaves out of the forest has reduced the amount of humus. “When a forest soil is exposed to desiccation and the fallen leaves are carried away by wind, the earthworms vanish, the soil becomes dry and hard, and the vegetation suffers. In acid soil (bog, heath) and dune, earthworms are wanting. Upon their presence or absence depends the occurrence of a humus soil or a raw humus soil in north temperate forest and heath. Con- versely they disappear upon the production of raw humus and humous acids. Even upon the growth of rhizomatous plants in the forest do they exert an action; their presence or absence causes a series of variations in the kinds of soil that corresponds to a series of variations in the plants clothing it.” 5 It is noted also that a covering of leaves in a beech forest has a great influence upon ground vegetation in that it supports mosses and other plants, produces humus, and provides food for animals living in the 1 W. J. Spillman, Renovation of Worn-out Soils, Farmers’ Bull. U. S. Dept. Agri. No. 245, 1906, p. 5. 2 E. Ebermayer, Lehre der Waldstreu, etc., 1876, p. 206. 3 E. Warming, (Ecology of Plants, Ox. ed., 1909, pp. 62-64. 4 C. S. Sargent, Manual of the Trees of North America, 1905, pp. 155-201. 5 E. Warming, CEcology of Plants, Ox. ed. 1900, p. 78. HUMUS AND THE NITROGEN SUPPLY OF SOILS 79 forest soil, among which earthworms are considered to be the most important.! One of the most important qualities of humus that affects its value to the soil is its natural porosity, which renders it very absorptive of gases, especially aqueous vapor. Dry humus swells up when wetted and the volume of weight increases in the ratio of 2 or 8 to 1; in fact humus stands first in this respect among the soil ingredients. Although in general the presence of the organic matter of plants increases the power of soils to hold moisture, some kinds of organic matter are known to cause a low water-holding power, as in certain California soils, a condition due to the peculiar and special qualities of the organic matter, which when extracted is found to have the properties of a varnish, repelling water to an extreme degree.? The density of natural humus as compared with ordinary soil is about 1:4. It is the lightest soil material and greatly promotes tilth, aeration, water supply, etc. Besides nitrogen, humus contains mineral plant food ingredients which are capable of nourishing plant growth. These mineral ingredients are probably made available to plants largely through the direct and indirect action of the humus compounds; for it has been shown that the richer the soil is in humified organic matter the more rapidly the mineral matter of the soil is made available for plant nutrition. With an increase of soil humus there is a correspond- ing increase in the amount of mineral plant food extracted from the soil by a 4% solution of ammonia such as is employed in the Grandeau method of humus determination (p. 82).* ORGANIC MATTER IN THE SOIL The difference between soil and a mere mass of sand or disintegrated rock is that soil contains some organic matter. Soil becomes arable and furnishes a medium suitable for the growth of higher plants when a certain amount of organic material has been accumulated in it from the growth and decay of lower plants. Animal remains, such as insects and worms, also have a prominent place as a source of soil organic matter. The accumulation of the remains of micro-organisms and of the vegetation which they modify is an important factor in the transformation of land waste into soil. The final product of these and other processes is a mix- ture of stuff fully as complex as the processes to which it owes its origin. 1 E. Warming, (cology of Plants, Ox. ed. 1900, p. 74. For a very clear and comprehensive discussion of the chemical modification of forest litter and its value to the soil, see E. Ebermayer, Die gesammte Lehre der Waldstreu mit Riicksicht auf die chemische Statik des Waldbaues, 1876. 2 Schreiner and Shorey, Chemical Nature of Soil Organic Matter, Bull. U.S. Bur. Soils No. 74, 1910, pp. 8-9. 3 E. F. Ladd, Bull. So. Dakota Agri. Exp. Station, Nos. 24-32, 35, 47. Quoted by Hil- gard, pp. 140-141. 80 FOREST PHYSIOGRAPHY The organic matter of both plants and animals is made up of protein, fats, and carbohydrates principally, but besides these substances there is a host of other compounds such as resins, hydrocarbons, and derivatives, that is, alkaloids, acids, etc. Furthermore, all living matter has as essential components some organic compounds that contain nitrogen, compounds which for the most part are those related to protein. When the complex molecules of proteins, fats, and carbohydrates break down into simpler bodies they pass through the same changes whether these changes are brought about through the agency of micro-organisms as in decay or through the agency of acids; and they are subject to still fur- ther decomposition through the same or other agencies. The secondary products are very numerous and of widely varying composition and structure, as well as chemical and physical composition and, properties, so that the final products of decay are very different under different conditions. When organic matter is contributed to the soil there is a continuous “building-down”’ process from the original complex molecule to sim- pler ones and these again to still simpler molecules, until in some instances the substances are reduced to their most elementary constit- uents.! All these chemical changes are affected by the temperature of the soil, the amount of air it contains and the amount of water, etc. Dry leaves, wood, and litter generally do not change in dry air or change very slowly; but if the ground is moist the process goes on rapidly.” In the absence of air bacterial transformations cease and the bacteria die. Extremely high or low temperatures likewise limit their activities (p. 86). Soit Humus AMOUNT AND DERIVATION The chief source of soil nitrogen, as we have already noted, is the organic matter in the soil, and this occurs toa large extent in the form of “humus.” The total amount of organic matter in ordinary soils is 2.06% for the soil and 0.83% for the subsoil, figures based on the analysis of thousands of samples from many different portions of the United States.* There is an average of 28 tons of organic matter in the soil per acre taking it to a depth of 8 inches, and 50 tons of soil and subsoil taking it to a depth 1 Schreiner and Shorey, The Isolation of Harmful Organic Substances from Soils, Bull. U.S. Bur. of Soils No. 53, 1909, pp. 9-11. 2 E. Ebermayer, Die gesammte Lehre der Waldstreu mit Rucksicht auf die chemische Statik des Waldbaues, 1876, p. 202. 3 Schreiner and Reed, Certain Organic Constituents of Soils in Relation to Soil Fertility, Bull. U. S. Bur. Soils No. 47, 1907, p. Io. HUMUS AND THE NITROGEN SUPPLY OF SOILS 81 of 2 feet. The amount of nitrogen is generally not far from 0.1%. In arid soils the average amount of humus is much lower, rarely ex- ceeding 1% and frequently falling to or below 0.30%; but the nitrogen content of the humus of arid soils is very much higher than in the case of humid soils. Woodlands and old meadows as a rule show a high humus content in their surface soils. The humus content of peat and marshland is also high. While figures for the amount of humus which may be derived from a given quantity of vegetable matter must vary greatly according to the conditions under which humification takes place, it may be said that in the humid regions roughly one part of normal soil humus may be formed from five to six parts of dry plant débris. This ratio is based upon the assumption that the average nitrogen content of plant débris is 1%. The ratio will vary according to the temperature, the degree of access of air and moisture, etc. In hot arid regions all vegetation may wholly disappear by oxidation at the surface of the ground, and the proportion of humus derived from the decaying vegetation of arid regions is in general very much smaller than that of humid regions where it is rather rapidly incorporated in the surface soil.’ The absolute amount of humus decreases rather regularly downward except in the case of depths that represent the maximum root develop- ment, at which level there is always a slight and often a notable in- crease in the humus content. Below this level decrease again takes place. The nitrogen percentage in the humus (which is to be distinguished from the total nitrogen content of the soil) exhibits a general decrease in the same direction, probably due to decrease in the amount of oxygen and a diminished rate of oxidation with increase in depth. The humus content of soils has a very close correspondence in some cases with the root development and is large in the surface layer and small in the subsurface layer. In other cases, while the decay of organic material takes place chiefly at the surface, active animal agencies may carry the organic remnants downward into the soil and effect a somewhat uniform distribution. If both root penetration and animal agencies are restricted by the compactness of the subsoil only a light surface layer of mold will be formed and what little humus is found in the lower soil layers will be derived entirely from the decay of a very limited number of roots. Cultivation or timber cutting when followed by exces- sive erosion prevents the accumulation of vegetable matter, from which 1 Schreiner and Shorey, The Isolation of Harmful Organic Substances from Soils, Bull. U.S. Bur. Soils No. 53, 1900, p. 26. 2 E. W. Hilgard, Soils, 1907, p. 128. 82 FOREST PHYSIOGRAPHY humus is chiefly derived, and allows the too rapid aeration and destruction of the humus. In general it seems necessary to keep the nitrogen percentage of soil humus at about 4% to insure satisfactory results, and for the:growth of grasses a nitrogen percentage in the humus of 1.7% is quite inadequate, no matter how much humus may be present. Different plants will accept this minimum, as might be expected from the differences of root habit, water supply, lime percentage, etc., which have an influence upon the rate of nitrification and of the leaching of nitrogen from the soil. If a moderate amount of moisture is present in the soil and there is in consequence a relatively free circulation of the air, and if earthy carbon- ates are present, especially lime, so as to neutralize the acids of the soil as fast as formed, fungoid and bacterial growths effect the steady humification of organic matter. Oxygen and hydrogen are eliminated in the form of water and carbon dioxide, and there is an increase in the percentage of carbon and generally of nitrogen. When once humification is completed oxidation affects mainly the carbon and the hydrogen, so that the nitrogen content may for a time rise steadily and reach very high figures. Under unfavorable conditions the conversion of organic material to soluble nitrogen may be so slow that a soil containing as much as 40% of unhumified organic matter may contribute during a single year but a small quantity of available nitrate. HUMUS DEFINED Originally the term Awmus had no special significance and was only a name for dark-colored vegetable mold; later it came to be applied to this mold material when incorporated to a greater or lesser degree in soils. The term now has a more restricted meaning, at least among soil chemists, and is applied exclusively to the dark-colored organic matter extracted from soils by dilute solutions (usually 5% solutions) of sodium or ammonium hydrate.' The method of determination is purely empirical and ascertains the existence of only a portion of the organic matter of the soil. Among agricultural folk and in general among foresters the term humus still retains its early meaning—partly decomposed organic material, dark in color, light in weight, and mixed with more or less mineral matter. The amount of humus in the soil is sometimes determined by means of dry or wet combustion in which process the humus is calculated from the carbon dioxide formed and the nitrogen gas is measured directly. Obviously this measurement includes the entire organic matter of 1 Methods of Analysis, Humus in Soil, Bull. U.S. Bur. Chem. No. 107, 1907, p. 19; see also C. A. Davis, Peat, 8th Ann. Rept. Mich. Geol. Surv., 1906. HUMUS AND THE NITROGEN SUPPLY OF SOILS 83 the soil whether humified or unhumified. To obtain the amount of actual functional humus in the soil (only humified organic matter can be directly nitrified) a solvent must be employed and discrimination between humified and unhumified material made. The accepted process is the method of Grandeau. By this method the soil is first extracted with dilute acid in order to set the humic substances free from their combinations of lime and magnesia and a subse- quent extraction is made with moderately dilute solutions of ammonia. After the evaporation of the ammonia solution the humus is left behind in the form of a black substance somewhat resembling soot. This is then weighed and afterwards burned and the ash weighed. The amount of functional humus is considered to be the difference between these two weights. The nitrogen content of the humus may be determined directly by substituting in this process potash or soda lye for ammonia water and determining the nitrogen by the Kjeldahl method in the filtrate.t The chemical composition of humus is indefinite because it itself is a variable mixture of substances of complex composition. It always contains more carbon and nitrogen and less oxygen and hydrogen than the substances from which it was formed. The following table shows the chemical composition of grass and of the top brown layer of turf in a peat bog, also the composition of peat of greater age, at 7 feet and 14 feet respectively. CHEMICAL COMPOSITION OF VARIOUS TYPES OF ORGANIC MATTER? Grass | Top Turf Peat at 7’ | Peat at 14’ Garbotisa:= 00 50.3 Sofas 62 | 64 Hydrogen...... aS | 5-4 ism 5 Oxygen..s.2.: 42.3 | 36 30.7 | 26.8 Nitrogen...... 1.8 | 0.8 2a 4.1 | It should be remembered in the inspection of this table as well as in the general consideration of humus that vegetable matter con- sists, in addition to carbohydrates, of other carbon compounds con- taining nitrogen, and in some cases both nitrogen and phosphorus, which all break down under bacterial and acid action into dark-colored substances called humus. In the process of humus formation the nitrogen-containing bodies resist the action of bacteria longer than the carbohydrates, so that the later the stage of decay the greater the pro- portion of nitrogen the humus will carry. It follows that during the gradual depletion of the humus higher and higher percentages of nitro- gen will be developed in that part not removed.’ Humin and humic acid are terms used in designating vegetable acids resulting from the decay of vegetable matter in the formation of peat, etc. The descriptions and formule for humic acid and related bodies 1 For a description of the Kjeldahl method see Wiley, Soils, 1906, p. 491 et seq. 2 A.D. Hall, The Soil, 1907, p. 42. 3 Idem, pp. 41-44. 84 FOREST PHYSIOGRAPHY are about as numerous as the number of investigators.!_ Certain elabo- rate experiments have resulted in the attempt to show that humic acid consists for the most part of an insoluble body of a protein nature, but this proves only that the humic acid examined by this investigator was either of a protein nature, a mixture of protein decomposition products, or probably both, together with some unknown body.” Humic acid, ulmic acid, ulmin, etc., are commonly used as if they were definite bodies of well-established composition, but this is not the case, as their very existence has never been satisfactorily demonstrated. No attempt should be made to ascribe formule to acids of so complex and variable a nature. As commonly written the formule for the group are as follows: 8 ULMIN AND ULmic ACID Carbone Geer psec wie tere anil eas, 67.1% le hefohgoyecelavann tian ony en ay Seer dN el 4.2 Corresponding to C4oH2s012 + H2O0 Oxy Pena tie sea ee eae ance 8.7 ! Humin AND Humic Acrp Carbone)... Uk Osean er WU esa 64.4% yd rogenta st ceuk euiete ieee tenes 4.3 Corresponding to CoiH24012+ 3 H20 Oxy General sesh cee ete Bila Crenic AcID Carbone een hci oes 44.0% Ely drog eater. deee getters hares aera 585 5 > INItrOR EMM egese/ eaierera eons eS 3.9 | Corresponding to Ci2Hi20s Oxy eM s et eee Aes Weeeseen nee 46.6 APOCRENIC ACID Carbonate ets ir cee einen 34.4% Fy dro Penky wen Ae tine pieere waa 35 = sane INitrogente sen eee eens. Be Pomesnoncing to CesHisdOre? Oxy Peni sa resee wines eaten barnes 39.1 The ulmic, crenic, and apocrenic acid groups are therefore names for exceedingly complex and unstable compounds as yet but very little understood. Although they have never been isolated and their char- acter definitely determined it should be remembered that there is reason for believing that they have at least a very short-lived existence in the soil. If they exist at all they pass quickly into the higher stages of oxi- dation and their final condition is COz; yet it is believed that during their supposed brief existence they may not only attack alkalies and alkali earths but also dissolve even silica.* Clarke asserts that they have a very appreciable solvent power and advance the decompo- sition of rocks, but notes that their constitution is but little understood.5 1 Schreiner and Reed, Certain Organic Constituents of Soils in Relation to Soil Fertility, Bull. U. S. Bur. of Soils No. 47, 1907, pp. 14-16. 2 Suzuki, Bull. Col. Agri. Tokio, vol. 7, 1907, p. 513. Quoted by Schreiner and Shorey in Bull. U.S. Bur. of Soils No. 53, p. ro. 3 G. P. Merrill, Rocks, Rock-weathering and Soils, 1896, pp. 189-1090. 4 Sir Archibald Geikie, Text-book of Geology, 3d. ed., 1886, p. 472. 5 F, W. Clarke, The Data of Geochemistry, Bull. U. S. Geol. Surv. No. 330, 1908, p. 400. HUMUS AND THE NITROGEN SUPPLY OF SOILS 85 Hilgard says that an acid reaction is characteristic of the soils of many woodlands, as is notable of the soils of the long leaf pine region of the South as well as of many deciduous forests in northern climates. He believes that in the course of time oxidation converts the natural neutral humin and ulmin into humic and ulmic acids capable of combining with the bases. Still further action is thought to result in the formation of crenic and apocrenic acids, which are readily soluble in water and in part form soluble salts with lime, magnesia, and other bases. “These acids act strongly upon the more readily decomposable silicates of the soil, and in the course of time may dissolve out, and aid in the removal, by leaching, of most of the plant- food ingredients . . . of the soil.’’1 It is conceived that this agency may be responsible for the almost complete absence of mineral plant food in the lower portions of peat beds and the subclays of coal beds. Besides the uncertain humus compounds of ulmic, humic, crenic, and apocrenic acids, there are others of similar derivation the action of which is not yet understood; among them are xylic acid, saccharic acid, and glucinic acid, and a black humus acid containing 71.5% of carbon and 5.8% of hydrogen. In the old forests of northern climates there is sometimes formed at the surface a partially decomposed peaty layer or vegetable remains which retards the full production of the land both while occupied by forests and for some time after being cleared.” In the case of such arrested humus development the soil becomes sour. The sourness is thought to be due to the presence of the above-named acids. Raw humus or unhumified organic material consists of incompletely decomposed plant remains, and may be found so rich in such remains from 50% to 60% organic matter as to be employed as fuel. It has free vegetable acids in abundance, earthworms can not penetrate it, and by itself it is of no value to plants. Raw humus appears in the forest in poorly drained places or in places exposed to wind, while ordinary humus with its earthworms, insects, etc., occurs in fresh, sheltered places. Ordinary humus or vegetable mold contains many fungi, besides earth- worms and insects, and is an excellent nutritive substance for plants: the rapid production of humus in the forest is therefore a kind of natural manuring. When ordinary humus in the beech forests of Europe is displaced by raw humus because of timber falls, etc., the beech is no longer capable of regeneration and disappears, being replaced by a heath.? Unhumified organic material has a potential value through the pos- sibility of its nitrification into active humus. It also lightens the soil by rendering it more pervious to air and water, and by progressive decay gives off carbon dioxide, which is the basis of carbonic acid so important in soil decomposition. 1 E. W. Hilgard, Soils, 1907, p. 126. 2 Muller, Naturliche Humusformen. 3 E. Warming, (Ecology of Plants, Ox. ed., 1909, pp. 62-63. 86 FOREST PHYSIOGRAPHY ACTION OF BACTERIA We have as yet only briefly mentioned the action of bacteria in rela- tion to soil nitrogen. Further discussion is required in order that the conclusions concerning the control of bacterial action may be adequately understood. Since the growth of bacteria is a large factor in maintain- ing the supply of some of the most important plant foods, the conditions which promote the activity of such bacteria are matters of serious interest to those engaged in the production and care of plants. Bacteria are plants that form the simplest group of fungi and are lacking in chlorophyll. They are very minute. The largest forms may reach a diameter of 0.008 mm. (0.000352 inch), and the majority are not more than 0.005 mm. (0.000197 inch) in diameter. It is believed that some bacteria are too small to be seen even with the most powerful microscope. Though they are of small size they are concerned with almost every phase of our daily life and overcome their apparent insignificance by their incredible numbers and ceaseless activity. A fertile soil has from 500,000 to to million bacteria to the gram, or from 15 million to 300 million to the ounce.!. In a drop culture of one cubic millimeter an experimenter found that one-tenth the total volume was composed of bacteria. In 24 hours, 48 generations will produce 281,500 billions of organisms.2. They are most numerous near though generally not at the surface, de- crease in numbers rapidly downward, and generally vanish at seven or eight feet. Water drawn from deep wells dces not show any bacterial growth. The functions and value of soil bacteria are variable, but the kinds that thrive in the soil are chiefly beneficial. Their action in the soil is affected by moisture, temperature, degree of comminution of soil particles, aeration, drainage, etc., besides which bacteria have associa- tive relations with each other whose reactions may be either bene- ficial or harmful. Some of them decompose dead plant and animal matter into simpler compounds, reconstruct various inert materials, and thus constantly renew certain elements in the soil and maintain its fertility. It should be remembered, however, that if the conditions of food supply and environment are unfavorable harmful groups of bac- teria may destroy the fertility of the soil.’ Humus is essentially a product of bacterial action, but this action should be carefully distinguished from the later action, called nitri- fication, which transforms the nitrogen of the humus into available nitrates. The formation of humus is accomplished mainly by the breaking down of carbon compounds, especially the carbohydrates of plant tissues, with the production of marsh gas or hydrogen, carbonic acid, and humus. With a surplus of air the humus-forming fermenta- 1K. F. Kellermann, The Functions and Value of Soil Bacteria, Yearbook U. S. Dept. Agri., 1900, pp. 219-226. 2 Grandeau, Ann. Sci. Agr., vol. 1905, p. 456. 8 For Winogradsky’s classification of nitrifying organisms see H. W. Wiley, Prin. and Prac. of Agri. Anal.: Soils, vol. 1, 1906, p. 557. HUMUS AND THE NITROGEN SUPPLY OF SOILS 87 tion is replaced by one which results in the complete combustion of the organic matter to carbonic acid. It is largely for this reason that more humus is found in a pasture or a forest than in a continuously tilled field. Most aerobic (oxygen-consuming) bacteria require for their well- being, or even in some cases their existence, some carbon compounds of nitrogen, and will begin to break down proteids and other nitrogen-con- taining materials. The products of their action, Fig. 9, are successively peptones (like leucin and tyrosin), eventually ammonia, and probably free nitrogen, but the formation of ammonia is the most characteristic effect of the bacterial fermentation of a proteid. When organic matter has decayed to the stage of such simple compounds as ammonia, nitric acid, carbonic acid, etc., it is no longer organic matter and much of it may escape from the soil altogether.’ Ammonia is not directly assimilable in the soil when delivered to it by the air or when occurring as the product of plant decay. The ammonia of air and soil is converted into nitrous acid by bacteria or by the oxi- dation produced under the influence of the catalytic activity of ferric hydroxide. The latter process takes place at a temperature of 15° to 25° C., and under its influence a certain amount of available nitrogen is developed in the soil independently of the activity of the nitrifying fer- ments. The conversion of ammonia into nitrates is, however, chiefly accomplished by two groups of organisms, (1) nitrosomonas or nitro- sococcus and (2) nitrobacteria. The action of the first group is limited to the formation of a nitrite; the action of the second group is the oxi- dation of the nitrite to a nitrate. Both groups are active only upon humus and its products, and are to be distinguished from the ammoniacal bacteria (such as Bacillus mycoides) that effect the reduction of the carbohydrates and the oxidation of proteid compounds to humus and ammonia. The latter have wholly different habits. In general the nitri- fying organisms require both organic and mineral substances for proper growth. Indeed some forms of nitrifying organisms have the power of subsisting wholly upon mineral substances.” The proper conditions for both groups of organisms are somewhat definite. A fairly high temperature, 75° F., is most favorable, and there must be a certain amount but not an excess of moisture present in the soil. If the temperature is low and water is present in excess, bacterial action may be incomplete in its effects or cease altogether. This is illustrated by the preservation of leaves, stems, and seeds for thousands of 1 Schreiner and Shorey, Chemical Nature of Soil Organic Matter, Bull. U. S. Bur. Soils No. 74, 1910, p. 45. 2 H. W. Wiley, Prin. and Prac. of Agri. Anal.: Soils, 1906, pp. 522-526 et al. 88 FOREST PHYSIOGRAPHY years in peat bogs, and by the well-known antiseptic properties of sour humus or peat. Free oxygen in large quantities is required, and there must be present a base or its carbonate, such as lime, with which the acids due to oxidation immediately unite. Sour soils exclude nitrify- ing bacteria through the action of the organic acids that have not been neutralized. The neutralizing salts favorable to bacterial development are not restricted to the carbonates. Potassium chloride acts favorably up to 0.3% but suppresses nitrification at 0.8%. Earthy and alkaline sulphates act favorably up to 0.5%. Among the latter gypsum is most beneficial and accelerates the process more than any other substance known. Common salt inhibits nitrifica- tion to a notable degree, and to this fact is due in great part the absence of nitrates in low- lying seacoast tracts. Arranged in the order of their value to the nitrifying process the various substances stand as follows, taking gypsum at 100% (after Pichard). Gay PSUn Seay aetna Oech pM a yeas sf ks RAT cee Pe ily fee NO teers AT OLGA Sodiessul plates eee sais te tama (ects oc tuk tire nina apne (UNG anh oral eee OS a Se 47.9 POtCASSIC Sul pl ates cy sacs seme seaeke te aus Oy rea Atte otis eee Se eB cee TR ES ae 35-8 Galcicrcarbomate cies ies Sirs) caste avscksics atuveetee ae ccs taal cane open coatee cca ae bat Rata 1323 Migenesic/ carbonate sain. rccsiais tunel cestevene) ale id Sola vostat cues cute Neues ae censor artiste ceniere TOPs The value of thorough aeration in nitrification is shown by the experi- ments of Deherain in which a cubic meter of soil was left undisturbed for several months while a similar mass was agitated in air once a week for the same period. At the end of the experiment it was found that the ratio of nitrogen content in the two samples was I: 70 respectively.! It has also been shown? that the brown substances of humus and analo- gous compounds are directly oxidized to some extent under the influence of air and sunlight, thus forming carbonic acid. The process is purely chemical, has no relation whatever to bacteria, and is rendered more effective by cultivation, principally through the better aeration thus effected. The succession of changes through which organic matter passes in the processes of nitrification and denitrification is shown in the accompanying diagram, Fig. 9. It presents in summary form the principal changes that have thus far been described. The immensely important conclusion has recently been established that all soils contain groups of protozoa which feed upon living bacteria and restrict their numbers, thus acting as beasts of prey.* Their preda- tory activities seriously restrict the limits of nitrogen production even : when the amount of organic matter in the soil is greatly increased. Happily a remedy has been found in heating or in treatment with anti- septics. Crop increases of 30% have been effected by a 48-hour treat- 1 E. W. Hilgard, Soils, 1907, p. 147. 2 Berthelot and Andre, Comptes Rendus Academie de Paris, 114, 1892, pp. 41-43. 3 A, D. Hall, The Fertility of the Soil, Science, n.s., vol. 32, 1910, pp. 370-371. HUMUS AND THE NITROGEN SUPPLY OF SOILS 89 Animals Nitrogen in the form of Proteid Plants Nitrogen in Free Nitrogen Nitrogen in the form of or Nitrogen the form of Nitrate Gas Ammonia Useless to Food for some Crops Most Valuable food for Nitrogen in the form of Nitrite Poison to Crops Fig. 9. — Diagram indicating the nitrogen changes in the soil produced by the action of bacteria. The arrows indicate the course of the changes which various groups of bacteria may produce in the nitrogen compounds of the soil. A, action of ammonifying bacteria which change organic nitrogen to ammonia; B, action of nitrifying bacteria which change ammonia to nitrite; C, action of nitrifying bacteria which change nitrite to nitrate; D, assimilation of nitrate by green plants; E, action of denitrifying bacteria which change nitrate to nitrite; F, action of denitrifying bacteria which change nitrite to ammonia; G, action of denitrifying bacteria which change ammonia to nitrogen gas; H, action of bacteria which change nitrogen gas into proteid nitrogen; I, action of bacteria which in symbiosis with leguminous plants change nitrogen gas into proteid nitrogen; K, action of bacteria which in symbiosis with certain non-leguminous plants change nitrogen gas into proteid nitrogen. 1 Yearbook, Dept. Agri., 1909, p. 222. go FOREST PHYSIOGRAPHY ment of the soil with vapors of toluene, chloroform, etc., followed by complete volatilization of these antiseptics. Analyses of the plant material so produced shows an assimilation of greater amounts of other plant foods as well as of nitrogen. It follows that the extra growth does not represent mere temporary stimulation but an absolute increase in the available stores of plant food. While great numbers of bacteria also succumb in the application of the remedial measures, some of them escape, and these, immune from attack, increase at a prodigious rate and almost at once increase the soil fertility." Several species of bacteria have the power of direct fixation of nitro- gen from the soil air. Some of the most important of these bacteria are Clostridium pasteurianum, Bacillus alcaligenes, Bacillus tumescens, Pseudomonas radicicola, Granulobacter and several species of Azoto- bacter.2, The abundance of these bacteria in the soil seems to indicate the measure of the natural nitrogen-recuperative power of the soil. In the coastal-plain soils the genus Azotobacter occurs only in the surface layer of a few inches; in the deep and almost exhaustless soils of certain sections of the West the same genus is found in active condition even down in the fifth foot below the surface. While most investigators attach considerable importance to the direct fixation of nitrogen, Hall considers that it is yet to be ascertained if the direct fixation of nitrogen by bacteria has any very important part in re-creating the store of uncombined nitrogen in the soil. An interesting and elaborate series of experiments by Lipman* makes it seem likely that Azotobacter chroococcum has the power of fixing atmospheric nitrogen when in symbiotic association with certain green algee with which it is commonly found and which develop with great rapidity upon limestone soils. It has also been shown by Lipman that Azotobacter vinelandit does not require symbiosis with alge to fix atmospheric nitrogen, but that a mixture of it and another bacillus (No. 30) caused a doubling of the nitrogen development. Certain bacteria have nitrogen-fixing power when in symbiotic asso- ciation with various legumes such as clover, etc., and accumulate large amounts of soluble nitrates which are ultimately used by the host plant. 1 There are also present in the soil anaerobic or denitrifying bacteria whose life functions are not dependent upon the presence of air since they are able to avail themselves of combined oxygen by reducing the oxides present in the soils. Some of them cause the reduction of nitrates to nitrites and finally to ammonia as shown in Fig. 9. Such reductive processes are carried on by bacillus denitrificans I, and occur chiefly in soils rich in organic matter or badly aerated. The result is a loss of nitrogen and a depletion of the plant food in the soil. 2 Yearbook Dept. Agri., 1909, p. 225. 3 A. D. Hall, The Soil, 1907, p. 171. 4 Rept. Agri. Exp. Station, New Jersey, 1903-1904. HUMUS AND THE NITROGEN SUPPLY OF SOILS get és y oe Cat rites Sete eee tA See Fig. 10. — Upper figure represents nitrous ferment prepared by Winogvodsky from soil from Cito Lower figure represents nitric ferment from the same source. (Wiley Soils.) or g2 FOREST PHYSIOGRAPHY The bacteria live as parasites within the roots of the host plant from which they derive their carbohydrate food supply. The presence of the parasites stimulates the development, by the root, of the peculiar swellings known as root nodules or root tubercles. These become filled with a bacterial mass consisting principally of swollen and ab- normal (hypertrophied) ‘‘ Bacteroids”’ having forked outlines, but in part also of bacteria which remained in their normal condition.! For a time the nodule increases in size and the bacteria continue to furnish a steady supply of nitrogenous material to the host plant. Ultimately the nodules cease growing, the Bacteroids degenerate, and their sub- stance is absorbed by the host. Normal bacteria, however, remain and provide for future reproductions. When the root nodules cease growing the larger group of bacteria, called also bacteroids, gradually collapse and are then depleted of their nitrogenous substance. The bacteria capable of this action draw their supply of food largely from the plant with which they grow in sym- biotic relationship. It follows also that when the plant becomes sea- sonably inactive the bacteroids also become inactive for a season; they are found only upon actively growing roots and not upon the roots of dead legumes. A fourth process of nitrogen fixation is by symbiosis, in which the plant involved is one other than a legume. Whether this process has any practical value remains to be determined, but bacterial nodules have been found upon at least one species of Alnus (alder), upon Ceano- thus americanus (red root, New Jersey tea), Ceanothus velutinus, Elaeognus argentea (silvery berry), Shepherdia argentea (buffalo berry or rabbit berry), Podocarpus macrophylla, and several genera of the cycads, though the nodules of the latter group of plants are quite different in some ways from the nodules of legumes.? One investigator concludes that the property of utilizing atmospheric nitrogen belongs to many other plants than the legumes.? He contends that in all the plants examined by him structures were found which are capable of absorbing it from the air and transforming it into the organic state. . The chlorophyll cell seems to possess this power to a high degree. It absorbs free nitrogen and transforms it into organic compounds. Certain organs, called producers of nitrogen, occur in the tender parts of very young leaves or their petioles. But the whole matter of such utilization by non-leguminous plants requires much further work before these broad conclusions can be accepted. 1 Sharburger, Noll, Shenck, and Karsten. A Text-book of Botany, 3d Engl. ed., 1908, p. 232. 2 K. F. Kellermann, The Functions and Value of Soil Bacteria, Yearbook Dept. Agri., Igo0Q, p. 226. : 3 Jamieson, Rept. Agri. Research Assn. of the Northeastern Counties of Scotland, 100s, p. 16. HUMUS AND THE NITROGEN SUPPLY OF SOILS 93 It has been fairly well demonstrated that forests are able to appro- priate free nitrogen by means of their trichomes.’ In a large number of chemical tests conducted in such a manner as, it is believed, to exclude all other sources of nitrogen than the atmosphere it was found that nitro- gen is always present in the trichomes. Among the forest trees that. have been tested are Acer campestre, Tilia europaea, Ulmus campestris, Sorbus Aucuparia, Fagus silvatica, and Abies concolor. ‘The process is one of unusual interest, for it adds one more to the very brief list of means by which nitrogen is made available as a plant food. This statement should not be taken to indicate that plants in general are capable of fixing nitrogen. The experiments of two Rothamsted (England) inves- tigators, Lawes and Gilbert, and of Boussingault in the middle of the nineteenth century are very conclusive in their determination of the inability of cultivated plants to fix atmospheric nitrogen except in symbiosis with nitrogen-fixing bacteria.2, The decay of forest litter and its transformation into nitrates by bacterial means are the chief processes that supply nitrogen in forested areas.’ ACTION OF FUNGI Contributing to the general store of humus in the soil are a large number of fungi among which are Penicillium, Mucor, Aspergillus, Oidium, etc.‘ They take a prominent part in the conversion of vege- table matter into black neutral insoluble humus, a process that is always marked by the formation of carbon dioxide. It is always found that both the fungous tissues and the humus resulting from their decay are notably richer in nitrogen and carbon than the higher plants. These organisms are found about decaying roots and plants, though they are not confined to this occurrence. Their tiny fibrils or hyphe are scattered through the ground much more thoroughly than even the tiniest root- lets with which they are associated. It is this thorough distribution that makes their presence so important in soil fertility, for it results in a correspondingly wide dissemination of the humus to which the fibrils contribute. Their life functions are dependent upon air, so that thorough aeration promotes their activity and increases their number; it follows that they will be most numerous near the surface and that 1 T. Cleveland, Jr., Forests as Gatherers of Nitrogen, Science, n.s., vol. 31, 1910, p. 908. 2 A.D. Hall, The Soil, 1907, p. 162. 3 Warming suggests mutual reactions between plants as possible sources of nitrogen. “Experience has shown that where Picea excelsa has been planted in Jutland it flourishes better in company with Pinus montana than without it. It is probable that in this case the mountain pine provides the spruce with nitrogen.” 4 A. D. Hall, The Soil, 1907, p. 182. 94 FOREST PHYSIOGRAPHY the nature of the soil drainage will determine their maximum depth or even their very existence. From the standpoint of the forest it is fortunate that these growths are not confined to decaying vegetation. They infest the roots of a large number of trees and shrubs and enable the latter to assimilate indirectly the decaying organic and inorganic matter that would other- wise be unavailable. They are in symbiotic association or codperation with pines, firs, beeches, aspens, and many other forms, all of which appear to depend very largely for their healthy development upon such association.! The degree of dependence of the host plant upon the fungoids that inhabit it is emphasized by Frank, who says that some host plants are so dependent upon this relation to obtain the necessary carbon from the humus that they possess no green carbon- assimilating foliage, a condition illustrated by the Neottia nidusavis or Bird’s-Nest Orchid found among beech underwood in England. The action requires close association of host plant and fungus, and in some cases the fungus penetrates the cortical tissue of the root or forms a sort of cap on the ends of the smallest and shortest rootlets.2 The association, when it results in the supplying of the host, plant with mineral substances, is especially characteristic of plants that grow in soils subject to drought or poor in mineral salts or rich in humus. In general those plants that have feeble transpiration capacity are well supplied about their roots with fungus through which they obtain food of many kinds from the soil. Such plants have also a very limited starch development in the leaf. Fungi may be harmful in their action when they rapidly transform the plant food into available form and produce such large amounts as to cause the excess to be removed by solution, so that a luxuriant but short-lived growth occurs, as in the ‘faery rings’? common in poor pastures.? 1 E. W. Hilgard, Soils, 1907, p. 157. 2 A. D. Hall, The Soil, 1907, p. 183. 3 Idem, pp. 184-185 and 219. CHAPTER | VIE SOILS OF ARID REGIONS GENERAL QUALITIES THE rainfall of arid regions is insufficient to leach out of the soil the salts that tend to form in it during the progressive weathering of the rock. In humid regions where the rainfall is abundant these salts are leached out and pass in very dilute form through springs and streams into the sea, and relatively small portions of the salts formed by weath- ering are retained in the soil and utilized by plants. It follows that when the degree of aridity is variable the salt of the soil will be variable in kind and amount and that in extremely arid regions practically the entire mass of salts contained in the rocks remains in the soil. We have already seen that arid soils are potentially more fertile than humid soils, for abundant rainfall in the latter case leaches the soil of valuable fertilizing elements that are easily soluble. In arid regions these elements are retained in the soil and many of them are very pro- ductive. To this advantage is added the greater number of bright clear days in arid regions and a maximum of insolation. The favorable features of arid soils are illustrated by the bench lands about San Bernardino, which are fertile, warm, equable in temperature, free from alkalinity, and, where water can be directed upon them for irrigation, among the most valuable agricultural lands in the United States.! The most striking characteristic of the soils of arid regions is the uniformly high percentage of lime and, as a rule, of magnesia, no matter what the underlying geologic formations happen to be. This condition is all the more apparent in the United States because of the general absence of limestone formations in the more arid portion of the country west of the Rocky Mountains and the wide distribution of such forma- tions east of the Rocky Mountains. No matter what the formation is, whether it is the granite of the Sierra Nevada foothills, the eruptives of the coast ranges of California and Oregon, or the great basalt sheets of Idaho and Washington, they are all alike in producing soils with a high lime content. It is largely to the high percentage of lime that the flocculated condi- tion of arid soils is due. This is one of their notable characteristics, 1 W. C. Mendenhall, The Hydrology of San Bernardino Valley, Cal., Water-Supply Paper, U.S. Geol. Surv. No. 142, 1905, p. 17. 95 96 FOREST PHYSIOGRAPHY and is related to the great depth of root penetration and easy tillage in arid lands. While in humid regions the average nitrogen content of soil humus is less than 5%, in the upland soils of arid regions the percentage rises as high as 22%, with a general average between 15% and 16%. This is probably due both to the ‘presence of the lime which keeps the acids that tend to form in the soil completely neutralized, and to the deficiency of vegetable matter which allows more complete nitrification than when an excess of such matter is present. The absence of an abun- dance of vegetable matter in the surface soil and of other qualities that differentiate it from the subsoil is a distinguishing character of arid- region soils. Subsoil and surface soil have no sharp line of separation either in fertility or general appearance or composition. The decom- posed state of arid subsoils, the absence of that rawness that character- izes humid subsoils, and the possibilities of their utilization are shown by the manner in which quick growths of yellow pine (Pinus ponderosa) appear in the placer mines of the foothills of the Sierra Nevada of California where the subsoil was thrown out years ago and became the surface soil. The timber growth is now of sufficient size to be used for mine timber and a second young forest is springing up on the red earth which once appeared as barren as the desert.! In the case of the relatively insoluble ingredients of the soil such as quartz or silica, the substance making up the greater part of sand, the humid region contains the larger amount, 84%, as compared with the 69% of the arid region where other substances exist in greater pro- portions. So that while sand of humid regions ordinarily consists of a collection of quartz grains with relatively clean surfaces, in arid regions it consists of a great variety of minerals in a partially decom- posed condition. Mixtures of fine and coarse particles at the surface are also more common in arid than in humid soils, due to the absence of thorough sorting in water. Thus the soils of arid regions often have uniform physical and chemical characters to a great depth. Scantiness of rainfall in arid regions effects a great retardation in the rate at which clay forms from feldspathic rocks and the sediments derived from them. This is shown in the distribution of two broadly different soil types in the eastern wet and western dry portions of the country. The soils of the Atlantic slope are prevalently loams and con- tain considerable clay; the soils of the arid region are generally sandy or silty with a small amount of clay unless derived directly or indirectly from clay or clay shales. 1 E. W. Hilgard, Soils, 1907. SOILS OF ARID REGIONS 97 The amount of phosphoric acid contained in the soils of arid regions is not different from that contained in the soils of humid regions. Phosphorus is a substance tenaciously retained by all soils and appears to be independent of leaching. On the other hand the leaching process has such a marked influence upon the compounds of the alkaline metals, potassium and sodium, which are readily soluble in water, that the ratios of their percentages show a marked difference in humid and in arid regions. The average ratio for potash is .216% to .672%, the ratio for soda is .140% to .420%. Potash occurs in greater abun- dance than soda in all soils because it is tenaciously held through reactions effected by zeolitic compounds in which soda is wholly or partially displaced when brought into solution with a potash compound, so that soda accumulates only where the rainfall. or drainage is in- sufficient to effect proper leaching; in such places it results in what is generally known as an alkali soil. Potash is therefore relatively abun- dant in arid soils and is one of the last substances added to them for increasing their productivity. They rarely contain much less than 1% of acid-soluble potash, occasionally rising as high as 1.8%. Among the most notable differences in the composition of arid and humid soils is not only the smaller amount of humus found in arid regions but also the relatively higher nitrogen content of the arid-soil humus. On the average, the humus of arid soils contains about 314 times as much nitrogen as the humus of normal soils and in extreme cases the amount may be 6 times as great, in which case the nitrogen percentage in the arid humus considerably exceeds that of the albu- minoid group, so that in arid regions a humus percentage which in humid regions would be considered quite inadequate may be considered entirely sufficient for all crop demands.'! In arid regions the substance that first requires replacement is phosphoric acid, the second is nitrogen. ALKALI SOILS For reasons already stated in connection with the higher percentage of soluble salts in arid regions, certain tracts may under special topo- graphic and drainage conditions develop alkalinity, a condition due to the presence of three compounds which usually form the main mass of the salts—the sulphate, chloride, and carbonate of sodium. Among these, calcium sulphate is nearly always present, magnesium sulphate (Epsom salt) is in many cases very abundant, and calcium chloride is present occasionally. The composition of a more or less typical alkali soil in California is shown in the subjoined table.’ 1 E. W. Hilgard, Soils, 1907, p. 138. 2 Hilgard and Weber, Bull. Cal. Agri. Ex. Sta. No. 82, p. 4. Quoted by Wiley. 98 FOREST PHYSIOGRAPHY TABLE SHOWING COMPOSITION OF ALKALI SALTS IN SAN JOAQUIN VALLEY Tulare County | Pesnles|| Near cia : Tulare Goshen Dit a | Lake Visalia | Lemoore Expm’t ate Tulare Station Surface | Alkali | Surface | Surface Alkali Alkali Soil Crust Soil Soil Crust Crust Soluble salts in 100 parts soil........... Tqoeale® pe eyaet | 0.83 T1260. \|visc eer | eee eee SPOtaSsiUMeSUlp haters tre eee ete ee) || eee eens (eennc cee Vics ceecbvaie oyci| ces eeey cee | renee ne aaa small aPotassiumenitratecs sy ansateui arisen Clore ilaie torre Heeien yous a0 [loses yeeets DS ed small Potassium:carbonaten(Saleratus)icce:cecee-|| tisieiete ie el) Siete ciacte cilieiiese eteuese 18.805 < |e Se eee ea eee eee Sodium sulphate (Glauber’s salt)........ 44.24 1.22 |312.30(m)| 13.4 chiefly 32.8 Sodium carbonate (sal soda)............ 32.98 88.09 |18.2 Ae 3.4 les ce 13.16 Sodium chloride (common salt).........| 16.74 TOO) Miler ake 4-4 little 31.16 aSodium:phosphates wien eee DIO 7 ilccas eer eee 0.22 TO. Ae a CS Ee eee Calciumysulphated(gy sum) ier se seer rereaeel cues ceccet oes lec careterets ces nue See en ee are little zy ste eee Magnesiumsulphates(Bpsomesalt) spec, «alanis cuele «tse elas Suess) 'asatebeleterens 8.1 moderate |......... Organichmattenter aor wee eee ei 1.97 Qe 2E = | 7 Sth oltchelevere coho tel| Mecaea serene 5.37 a Very generally present, but not always in quantities sufficient for determination. Alkali lands are so widely distributed in the desert regions of the world! that the problem of their improvement and utilization for agri- culture is of great importance. These lands when properly treated have great fertility on account of the many soluble plant foods in them, and they may in many places be turned from waste lands to fertile oases. Their natural vegetation is of little value except in a few cases, as the salt bushes and wild clover of South America and Australia, which form valuable pasture. Considerable areas of alkali lands are either destitute of vegetation or bear resistant growths of little value as forage. The effects of sodic carbonate on plants grown in alkaline regions are seen in a scant leafage, short growth of shoots, and a deadening of the roots. The cortex assumes a brownish tinge just above the surface in the case of green herbaceous stems, and, in the case of trees, the outer bark assumes an almost black tint and the green layer underneath turns brown. The maximum injury is usually at or near the surface, where there is a maximum accumulation of salts, due to evaporation at the surface. The vertical distribution of the alkali salts in a California soil is shown in the diagram below, Fig. 11.* Certain native plants that live upon alkali soils have adapted their root systems to a very interesting condition. Figure rr shows that down to a depth of 15 inches there is practically no alkaline content; and it is within these 15 inches of soil that the native plants develop their roots 1 The total area of the arid lands of the world computed from the total area of interior basin drainageis given by Sir John Murray as 1114 million square miles, or one-fifth of the total land surface of the globe — The Origin and Character of the Sahara, Science, vol. 16, 1890, p. 106. 2 E. W. Hilgard, Soils, 1907, p. 432. SOILS OF ARID REGIONS 99 and develop their growth. The bulk of the salts accumulate at the greatest depth to which the annual rainfall of seven inches reaches, where it forms a hardpan. It is above this hardpan that the seeds of the shallow-rooted plants germinate and extend their roots. The soil mois- ture of the surface layer is so thoroughly consumed by the plants that no alkali is brought up from below by evaporation and the life cycle begins the following season. It is in this manner that the luxuriant vegetation of the San Joaquin plains is maintained except where occa- sional alkaline spots occur. The horizontal distribution of alkali is variable and the location of the salts changes from year to year, espe- cially in irrigated lands, so that the cultivation of alkali lands, the de- termination of their position, etc., must be carried on with great care. Amounts of Ingredients in 100 of Soil 0.02 .04 .06 .08 .10 .12 .14 .16 .18 .20 .22 .24 .26 .28 .380- .32 .34 .86 .38 .40 .42 .44 .46 .48 .50 .52 .54 ees 12 1 Foot i——} eeerae| ene eee noe WS a | ass panel Salt: (ut the- Ss Vs se HES be SS x : x Odic Carbondig T—-~ 1 ai Se : 4 Fig. 11. — Amounts and composition of alkali salts at various depths in black alkali lands covered with native vegetation. (Hilgard, Soils.) Two types of alkali soils are noticeable: the one is due to the presence of carbonate of soda and is called black alkali; the other is due to the presence of the sulphates and chlorides of sodium, and is called white alkali. The latter is much milder in its effect on plants. In California, outside the main valleys no important amounts of alkali salts are found at depths exceeding four feet. The total amount found in alkali lands which show saline efflorescences at the surface in the dry season is from one-tenth of one per cent to as much as three per cent of the weight of the soil taken to a depth of four feet. Alkali lands also have a high lime content and a high potash content — higher than the average amount of phosphates; while nitrates are usually scarce or altogether TOO FOREST PHYSIOGRAPHY absent, though nitrates may occur along the margins of the alkali spots.! In sloping valleys or basins where the central lowest portion receives the salts leached out of the soils of the adjacent slopes, occur belts of variable width in which the alkali impregnation may reach to a depth of 10 or 12 feet. Such areas are, however, quite limited and irreclaim- able, and the predominating ingredient is usually common salt, as is illustrated in the Great Salt Lake basin of Utah, in the Antelope and Perris valleys of southern California, and the Yellowstone Valley near Billings, Montana.? The salts of alkali lands are not permanent in their vertical position but follow the movement of the moisture, descending in the rainy season to the lower limit of the absorbed rainfall, and re- ascending in response to surface evaporation, so that at the-end of the dry season a saline efflorescence may occur or the entire mass may be found within a few inches of the surface. Carbonate of soda exercises a puddling action on the soil, destroys its crumb structure, and renders it almost untillable and impervious. It also tends to form a tough impervious hardpan as resistant to roots as to implements. Hilgard has shown that the proper treatment of alkaline soils is leaching (after treatment with gypsum in the case of black alkali), together with subdrainage. Flooding alone is not sufficient, for if the process is carried on generally the alkali spots grow larger to the destruction of adjacent lands. If codperative subdrainage is carried out the salts may be entirely removed by an excess of irrigation water or be carried down to so low a level as to have no injurious effect upon plants. The amount of alkali in the soil may be diminished also by cultivating plants that take up considerable amounts of salt, —a notable property of the greasewoods (Sarcobatus, Allenrolfea), which contain from 12% to 20% of alkaline ash. When grown upon the land and then cut and removed, such plants will markedly diminish the amount of alkali in the soil. A few such salt-consuming plants are fit for pasture, such as the Argentine plant (Atriplex chachiyuyun) and the Australian salt bushes (Aériplex halimoides), Vesicaria and Lepto- carpus, and a Chilean plant (Modiola procumbens). The results of the reclamation experiments of the U. S. Bureau of Soils are surpris- ingly good and indicate the range of possibilities in regard to the use of alkali lands.* Tracts originally covered with a white crust of alkali and 1 E. W. Hilgard, Soils, 1907, pp. 430, 444, 448. 2 Farmers’ Bull. U. S. Dept. Agri. No. 88, 1899. 3 For specific descriptions of the localities where alkali soils have been experimentally improved and reclaimed, see C. W. Dorsey, Reclamation of Alkali Soils, 1907, and Reclama- tion of Alkali Land in Salt Lake Valley, Utah, Bull. U.S. Bur. Soils No. 43, 1007. SOILS OF ARID REGIONS IOI supporting a scanty growth of greasewood were sweetened by flooding and drainage and then sown to alfalfa, various vegetables, grains, etc., with very beneficial results. The accumulation of mineral salts at or near the surface has given rise to the formation in most arid regions of a characteristic deposit known as “caliche.’’ It often consists of a variety of substances, but the most common constituent is nitrate of soda. The greatest deposit of this sort is found in Chile in the province of Tarapaca, where nitrate of soda is produced on a very large scale, practically the whole of the world’s supply being derived from this desert region. Smaller tracts are found in many other deserts, notably in the Southwest as in Death Valley, Cali- fornia, but the scale of production in all these cases is decidedly limited. In general the layer of caliche is covered with a deposit of earth from a few inches to a few feet in thickness. It may consist in part of wind- blown material, in part of water-laid material deposited since the caliche was formed. The largest beds of caliche are probably due to the crys- tallization of mineral salts from bodies of water which have disappeared either through a change of climate or a change in the level of the land or both. The origin of this class of material is, however, still in doubt, and although the result is closely allied to aridity it is not yet clear what combination of arid conditions with topography, drainage, and chemical character of rock and vegetation brings about its existence. CHAPTER VIII SOIL CLASSIFICATION Ir has been generally agreed among soil investigators that because of the predominating influence of physical characteristics in soil fer- tility, physical and not chemical qualities shall be made the basis of soil classification. This decision is strengthened by the immemorial cus- tom among agricultural folk of designating soils principally by their phy- sical character. The terms sand, gravel, clay, etc. (or their equivalents), are common non-technical words which convey a fairly definite meaning the world over. When, however, a soil is to be scientifically investi- gated, its characters strictly defined, and its value and its needs formu- lated, somewhat precise terms must be employed, careful experiments conducted, and conventional symbols devised which have stricter meanings than the colloquialisms of the farmer. Hence a relatively refined classification has been elaborated, based primarily upon physical character. In examining the accepted classification we should not lose sight of the fact that while the forester must acquaint himself with it in order to make the literature of soil investigators serve his purpose, he generally requires for ordinary field work a somewhat rougher scheme of classification. Gravel, sand, silt, clay, and peat or muck are the main types he is required to recognize, modifying his choice of terms by mention of such secondary qualities as the soil of a particular locality exhibits. He will also be required to distinguish between various grades of gravel, sand, etc., as coarse, medium, and fine, and it is obviously to his advantage to employ for these subdivisions the basis employed by soil specialists, in so far as this is possible. In determining the sizes of soil grains a number of methods may be employed; the three principal ones are (a) sieving the soil samples, (b) elutriating them, and (c) separating the various grades by the subsidence method. (a) The sieves used for soil analyses have round holes of carefully determined diameter. The unit employed may be fractions of an inch, but the smaller units of the metric scale (millimeters) are decidedly pref- erable both because of their international acceptance and their easy use in computations. The soil samples are sifted after being rubbed so as to destroy the lumps or soil crumbs composed of both fine and coarse I02 SOIL CLASSIFICATION 103 material that behave in some respects as large individual particles the size of the lumps. Separation may be more easily accomplished by playing water on the sieve; without it the clay particles and even the silt particles tend to cling to the sand as soon as the grain sizes in the latter fall much below 14 mm. (6) The elutriator, Fig. 12, is an instrument employed to separate soil grains of different sizes (after removal of the clay a NO by subsidence) by an as- A cending current of water Se at various fixed velocities. : The soil grains are carried ano off in exact conformity to their several sizes or volume Welsntcmerkhe maintenance- 0-465) eet Ni ls b of the current at a fixed Pre see IaH velocity for a long enough period will result in the ia practically complete _ re- V7] moval of all grains below a certain size. The different cry velocities are adapted to Fig. 12.— Elutriator (Hilgard’s) in position for soil analysis. certain desired grain sizes, and the volumetric determinations that follow elutriation form the basis for classification. (c) The subsidence method is based upon the assumption that if a soil sample is thoroughly mixed in water the different grain sizes will settle according to their weights. A successful outcome requires the removal of the clay by repeated sedimentations of the non-clay material, the decantation of the water in which the suspended clay particles are held, and the final sedimentation of the coarser grades. The necessity for the removal of the clay is due to the greater viscosity of the water in which the clay is suspended and its interference with normal and accurate sedimentation. The clay itself is determined from the several clay waters by evaporation of the water. Precipitation will not suffice, for the finest colioidal clay will not subside for years, —a condition thought to be due to a change in its physical and possibly in its chemical nature.! A defect of the subsidence method is the impossibility of abstracting all the clay, a defect the more serious because of the high importance of the 1 W. H. Brewer, On the Suspension and Sedimentation of Clays, Am. Jour. Sci., vol. 20, T8853) Dp: L- 104 FOREST PHYSIOGRAPHY clay constituent of soils (p. 35). Some clay particles are invariably carried down by heavier constituents and deposited with them. A similar result does not occur in elutriation because of (1) the agitation (which prevents flocculation into heavy aggregates) of the ascending current and (2) the grain sizes are expelled in reverse order, the finest first, and so on. PURPOSE OF A SOIL ANALYSIS The physical analysis of a soil is not alone for the purpose of finding a name for it in the series proposed in the table, Appendix A. This is the least of its purposes. It aims, in addition, clearly to present the controlling constituent among the soil grains. Soils are not generally composed of grains rather equally distributed among the several sizes. Some particular size usually predominates, and gives the soil its strongest individual character. This is clearly illustrated by the Volusia soils spread over some 10,000,000 acres of western Pennsylvania, southern New York, Ohio, and West Virginia. They are poor in their present condition, yet nothing in their chemical nature suggests poverty. Their unproductiveness is due largely to improper drainage, for their physical composition is such that the natural drainage is not adequate; acid therefore forms and accumulates and makes the soil sour. Their im- provement is not to be sought in the use of fertilizers alone, for the fine-grained fertilizers by themselves would still further clog the soil pores. It is suggested that adequate drainage and deep aeration would prove remedial.! : The remedies proposed are particularly interesting because they are among the relatively small number which the forester finds it possible to apply to economic advantage over large areas. The forester can not fertilize the ground by the application of manures or mineral fertilizers; the very scale of his work makes it impossible to do more than enable a forest soil to improve itself by encouraging processes already in action. This he is able to do in many ways. Proper cutting and seeding or replanting are almost always possible in both the physical and the commercial sense of the term, and by maintaining shade the drying of the surface soil, in which seedlings make their first growth, is prevented. Proper drainage is also feasible and is at once a means of partly control- ling the kind of growth and the rate of growth. It is fundamental in drainage to ascertain the nature of the subsoil as well as the soil. If the subsoil is open and porous it is likely to be dry; if clay forms the subsoil it will probably be wet and by capillarity 1 M. E. Carr, The Volusia Soils, Bull. U. S. Bur. of Soils No. 60, 1909, pp. 21-22. SOIL CLASSIFICATION 105 supply the surface soil with water in the dry season. Not only for this reason, but also because so important a part of the root systems of trees occurs in the subsoil, is it necessary to secure data as to its nature and its effects upon the surface soil in different seasons. DIFFERENT BASES OF SOIL CLASSIFICATIONS It is unfortunate that no general scheme of classification has been universally adopted by soil physicists. In Appendix A the scheme of the U. S. Bureau of Soils is described because the literature of this Bureau is now both extensive and valuable and reference to it indis- pensable on the part of any one beginning the study of the soils of a given region. Nevertheless it should be noted that many other bases of subdivision of soil types are in vogue. It has been proposed by Hilgard and others, that all constituents of soils that are too large to pass through a sieve with meshes 0.5 millimeter in width should be called the “soil skeleton” and that the remaining constituents that pass through the sieve should be called “fine earth.” He regards the fine earth as having a special relation to plant life as food material and through its physical attributes. For the purposes of ecologic studies Warming distinguishes six different kinds of soil as follows !: Rock soil. Clay soil. Sand soil. Humus soil. Lime soil. Saline soil. Soils may be distinguished also by classes, as rigid, stiff, mellow, lax, loose, and shifting, in order to express various grades of compactness. Hilgard suggests the use of broad types such as sand, clay, and humus. Considered in reference to their origin, soils may be classified as sedentary and transported. A sedentary soil may then be either residual where it remains upon the rock from which it was derived or colluvial where it is subject to slow down-hill movement on hill slopes. Transported soils may also be divided into alluvial where the soil is deposited on bench lands or flood plains, and eolian where it is de- posited by the wind. The most common classification is based on tex- ture, as gravel, sand, silt, clay, and their subdivisions and derivatives. Again, soils may be called humid or arid according as they are formed in one or the other of these climates. Or a soil may be classified according to its chemical properties. Thus, we have a lime soil with its lime-loving vegetation, or a magnesium soil, or a soil exceptionally 1 cology of Plants, Ox. ed., 1900, p. 60. 106 FOREST PHYSIOGRAPHY rich in potash or gypsum. A classification based on the distribution of vegetation is often helpful but the soil requirements of plants are not rigid except in the case of extreme types of soils and plants. Doubtless the physical classification has the widest practical impor- tance since physical features most commonly have a fundamental control over vegetation and are the most unchangeable. The ideal classification would be adaptable to, and would take cognizance of, all important soil characters. It is at least certain that no single scheme is applicable to all kinds of soils in all kinds of climates.! With this brief suggestion of the nature of other classifications in mind the student is referred to Appendix A for a description of a suggested outline of a soil survey with such description of the field methods re- quired as bear on soil studies practicable in forestry. 1 Hilgard and Loughridge, The Classification of Soils, Verh. der II Int. Agrogeologen- conferenz, Stockholm, rorr. PART TWO PHYSIOGRAPHY OF THE UNITED STATES CHAPTER IX PHYSIOGRAPHIC REGIONS, CLIMATIC REGIONS, FOREST REGIONS INTRODUCTION PHYSIOGRAPHY is indispensable to the environmental study of or- ganisms of every kind, whether trees, or men, or bacteria. Soil, topog- raphy, and climate are positive forces in the development of forests and the harvesting of forest products. They underline the main possi- bilities as well as the main limitations of nature. We have already seen that soil in at least small amounts is a necessary condition of tree growth; of the same order of importance are the facts that the broader forest distributions depend upon climate, while the accessibility of for- ests depends to a large extent upon topography. It is doubtful, for example, whether some of the best timber of the Sierra Nevada will ever be harvested because of topographic difficulties — steep canyon walls, sharp spurs, and remoteness from transportation lines. Climatic conditions exclude trees from the larger part of the arid and semi-arid West and from the higher, colder, and windier parts of the western mountains. Forests are excluded also from great areas of bare rock outcrop in regions of glacial denudation or from soils rendered infertile by extreme physical or chemical properties. The forester must take account not only of these relations but also of the larger relations of forests to stream flow and soil erosion. Each soil type has its own peculiar water-holding or water-shedding capacity, each topographic province has certain slopes upon which either agriculture or forestry can or can not be conducted, each natural region has its own climatic possibilities and restrictions, a study of which enables the forester to improve natural conditions and repress harmful forces to the benefit of mankind. It is our purpose in these pages to present the physical basis of forestry in the United States. No attempt is made, however, to discuss 107 108 FOREST PHYSIOGRAPHY either regional ecology or the principles of ecology. The ecology of the forest is regarded as a subject which possesses a body of facts and laws of its own. The single object of this book is to acquaint the forester with the: geographic basis of his work, with such references to the forest as point the direction of his more special subjects. This attitude should be appreciated here, lest the organization of the following chapters be misunderstood. For example, a forester requires a certain group of physical data in developing, let us say, the forests of the Black Hills. It is our object to discuss not the silvicultural or lumbering methods best adapted to the Black Hills and the relations of these to the physical geography of the region, but to make the student so familiar with the geography that a knowledge of it may be assumed when he begins a study of the forestry. PHYSIOGRAPHIC REGIONS The description and explanation of any large and varied portion of the lands proceed naturally by subdivisions, each subdivision embracing a tract in which the topographic expression is in the main uniform. Since uniformity of topographic expression is achieved only when geo- logic structure, physiographic process, and stage of development in the geographic cycle are also uniform, each subdivision has an essential uniformity or unity of geologic and physiographic conditions. It is customary to speak of each subdivision as a natural region or province, and to bound each region by lines which represent the limits of unity. In some cases the boundary lines are very precisely located, as along the eastern edge of the Allegheny Plateau in Pennsylvania, or the western edge of the Colorado Plateaus in Arizona, where great scarps mark out sharply defined borders; in other cases the transition zone between provinces is relatively wide and has characteristics which par- take of the nature of both adjoining provinces, as between the Columbia Plateau of southern Oregon and the Great Basin south of it. It must not be supposed that the idea of physiographic unity is applied in an absolutely rigid manner. Some of the physiographic provinces appear to have great topographic variation and but little unity. In such cases it appears at first sight impossible to group the forms in a rational manner until soil, climate, topography, and geologic structure are all examined, when prevailing or group characters always become apparent. Exceptions to group qualities are often observed, and in some cases these are of great importance, but on the whole they affect only the minor physiographic features. Thus the northern Rockies of Montana and Idaho are in striking contrast to the high plains on either PHYSIOGRAPHIC, CLIMATIC, AND FOREST REGIONS 10g side, — the Columbia lava plain of Washington on the west and the Great Plains of Montana on the east. This contrast between a belt of high, rugged mountains and gently rolling, bordering plains forms a primary basis for subdivision. Nevertheless the traveler in crossing the northern Rockies finds the landscape changing continually. Everywhere the ranges trend in the same general direction, but the kind of rock, the structure and hardness of the rock, and the kinds of dissection affecting the range masses vary from point to point; everywhere the major valleys have a distinctive trough-like cross section, but the minor valleys are of many varieties. No two ranges, therefore, are alike in detail, but all are alike in being rugged and high, while the bordering lands are smooth and low (relatively). Some of the mountains are dissected plateaus, as the Clearwater Mountains of Idaho; others are dissected anticlines and synclines, as the Lewis and Livingston Mountains of Montana; yet they are all alike in being deeply dissected. The broad similarities among the features of a physiographic province are frequently expressed in the name of the province, as “ Prairie Plains,” or “Colorado Plateaus,” or ‘Great Basin”; yet the Prairie Plains are locally rough, as where high and irregular morainic belts cross them, the Colorado Plateaus are diversified by volcanic mountains like Mt. Taylor and San Francisco Mountains, whose summits reach above timber line, and the Great Basin consists of many independent basins broken by, and mutually separated by, mountains of marked height and ruggedness. The dissimilarities among the features of a region may be classified and grouped by subregions. The Older Appalachians (Chap. XXVIII) for example have certain very distinctive and uniform features over a great belt of country from Maine to Alabama, such as their great geologic age, their highly complex structure, their prevailingly crystalline character, and the tremendous erosion which they have suf- fered. But throughout the region marked dissimilarities also occur which require recognition. The scenery about Asheville, North Carolina, is quite unlike that about New Haven; the mountain basin in the first case falls in the Appalachian Mountains, a subdivision of the Older Appalach- ians, the valley lowland of the second case lies in the glaciated New England subdivision of the same province. Many of the physiographic regions of the United States are of great size. The Great Plains are as extensive as European Russia; the Lower Alluvial Valley of the Mississippi is at least half as large as Italy; the Alps are less extensive than that part of the Rockies south of the international boundary. Many individual topographic features are developed on a large scale. Hurricane Ledge, an eroded fault IIo FOREST PHYSIOGRAPHY scarp in the Colorado Plateaus, is 700 miles long, the Great Valley of the Newer Appalachian region extends with but local interruptions from Alabama to Quebec, the Grand Hogback of western Colorado is a bold escarpment of erosion which extends unbroken for over 200 miles. In addition to the great size of individual provinces and features is the great variety of physiographic features which the country exhibits. The physiography is also in many respects unique. Exploration has not revealed anywhere else on the earth structures and forms like those of the Colorado Plateaus, either in respect of scale or perfection of de- velopment. The Newer Appalachian ridges and valleys are so perfectly developed that “Appalachian structure and topography” has become a technical phrase. The till sheets of the upper basin of the Mississippi, in the heart of the continent, are so extensive and their succession so complete that the history of the glacial period was first worked out to partial completeness in America. The Columbia lava flows constitute one of the few really great basalt fields of the world. We should ascribe to the great variety of physical conditions in the United States no small share of the general interest in American forestry, for the forests and forest problems are almost as varied as the relief upon which, either directly or indirectly, they so commonly depend. The vital relation of the forests of the arid West to the general welfare of the people has given western forestry a scientific interest not exceeded elsewhere. Man’s control of the desert had its beginning in his control of water. At first that control was in the nature of art; now it is in the nature of science. As much care was at first bestowed upon the ritual of rain as upon the construction of a dam; we now study the laws of rainfall, measure its several dispositions, exercise control over it in measur- able degree, relate cause and effect, and know the limits of irrigation enterprises before they are begun. Throughout the West the problem of water control has been found to be also a problem in forest control and grass control. Irrigation, forestry, and grazing are parts of one scientific problem. Of lesser but still of great importance are the rela- tions of run-off and forests in the humid East, where over large areas a mountainous relief diminishes the value of the land for agriculture and increases its value in relation to forests and stream flow. Mountain influences are extended out upon the plains, where resides a dense agri- cultural population whose commerce and towns and fertile valley lands are often largely dependent upon the behavior of the mountain-born streams. PHYSIOGRAPHIC, CLIMATIC, AND FOREST REGIONS III CLIMATIC REGIONS The determining factors of climate are chiefly latitude, the relative distribution of land and water, the elevation of the land above the sea, and the prevailing winds; the most important climatic elements are’ temperature, moisture, wind, pressure, and evaporation. The climatic elements are shown graphically on the accompanying series of maps for North America. Among them temperature and rainfall are of most importance in relation to forests, and a brief discussion of these elements of our climate is therefore presented in this chapter. In Plate I, rep- resenting the climatic provinces which correspond with the life regions of Merriam,! they are combined in such a way as to show their effect upon the life of the continent. ss ol & 4 AAD Ni y Fig. 13. K6ppen’s classification of climates in relation to vegetation. Tracts enclosed by broken line have distinct dry seasons (rain probability x SSS > SY t ¢ SS ¢ + Ss RS y c t x t e y t x t x SS Fig. r9. —Mean annual precipitation in the United States reduced to inches of rainfall. Note the Adirondacks, White Mountains, and Black Hills islands. (U.S. Geol. Surv.) California and San Francisco. Toward the south it continues to de- crease and falls to a minimum of less than ro inches at San Diego, in the horse latitude belt of light uncertain winds between the westerlies and the trades. The Pacific coast thus exhibits a range in rainfall of about 100 inches, The main Pacific coast valleys, embracing the Great Valley of Cali- fornia, Salton Sink, the Willamette Valley, and the Puget Sound depres- sion, have a much lighter rainfall than the rain-obstructing Coast Ranges, since they lie in the lee of the latter. The valley of southern California is an extremely dry desert. In the Great Valley the rainfall varies from about ro inches at Fresno in the south to 25 inches in the north; the PHYSIOGRAPHIC, CLIMATIC, AND FOREST REGIONS I1Q 105? 101° T 7 109’ 1057 Wy? 3 3% go? Site 81? Fig. 21. — The average annual humidity of the air in the United States. 120 FOREST PHYSIOGRAPHY rainfall of the Willamette Valley varies from 25 inches in the south to 45 inches in the north; while the Puget Sound region has an average rainfall of about 45 inches. The precipitation increases rapidly east- ward as the winds ascend the western slopes of the Sierra Nevada and the Cascades. It reaches a maximum of about roo inches in Wash- ington and Oregon, and from 4o to 8o inches in California at elevations between 3500 and 5000 feet. Beyond this point the precipitation Fig. 22. — Percentage of annual rainfall received in the six warmer months, April to September inclusive. diminishes again toward the summit and becomes insignificant at the eastern base of the mountains. The great height and continuity of the Sierra Nevada and the Cascades cause these mountains thoroughly to obstruct the westerly winds in respect of moisture, with the result that great stretches of country east of them are arid wastes. Where ranges of exceptional height occur in the country east of the Cascades and the Sierra Nevada the rainfall may exceed 25 inches a year, but by far the greater part of the region has less than 12 inches a year. Southwestern Arizona and southern Cali- fornia are the driest regions in the United States, and the rainfall of the lowlands is almost wholly confined to the winter months. In the rain shadow of the Sierra Nevada the mean annual rainfall is between 5 and 6 inches and locally as low as 3 inches. It is characteristic of the region PHYSIOGRAPHIC, CLIMATIC, AND FOREST REGIONS I2I that occasionally some portion of it receives a: rather abundant rainfall. About once in six years the greater part of the rainfall of a locality comes in a single month. In 1891, for example, 2.5 inches or 93% of the annual rainfall of the lower Colorado Basin fell in January. In January, February, and March, 1905, 8 inches of rain fell at Yuma, Arizona, whereas the average annual rainfall of that place is but 2.7 inches. The Rocky Mountain region rises to such a height as to provoke a heavier rainfall than the plateaus and basins on the west, though the remoteness of the region from the sea gives it a much lighter rainfall than occurs in the southern Appalachians of far lower elevation, or the relatively low Coast Ranges of Oregon. During the winter the western windward slopes of the Rockies are more heavily watered than the east- ern, but the reverse is true for portions of the system during the spring and summer. In New Mexico and western Texas the mountains of the Trans-Pecos region have a greater rainfall than the surrounding plains and basins, but it is never heavy in an absolute sense. It comes in July and August and appears to be rather evenly distributed on all sides as if due to local updraft of air from plains to mountains. Nowhere does more than 50 inches of rain occur in the Rockies and the average amount is far smaller. The rainfall is very unevenly distributed, as might be expected owing to the irregular trends of the intermont basins and mountain ranges and the variable dispositions and heights of the mountains. The maximum precipitation occurs probably in northern Idaho; the minimum is between 6 and 8 inches and falls in San Luis Park in south-central Colorado. Eastward of the Rockies as far as the Atlantic coast the topographic features lack great height, hence the rainfall distribution is controlled chiefly by the frequency and direction of movement of the rain-bearing cyclonic storms. The greater height of the Unakas, Great Smokies, and associated ranges in western North Carolina and South Carolina and northern Georgia cause their rainfall to exceed that of any other region east of the Pacific mountains. It is more than 70 inches a year. The rugged and high eastern portion of West Virginia, the Adirondacks, and the White and the Green mountains are other centers of heavy rainfall that owe their influence upon climate to their greater height. A heavier rainfall depending not on elevation but on nearness to the sea and to position within the track of frequent cyclones occurs in southern Louisiana and Alabama, — 60 to 70 inches. The Great Plains region has a diminished rainfall owing to its position in the rain shadow of the Rockies and its remoteness from the sea. 22 FOREST PHYSIOGRAPHY Fortunately such rain as falls comes chiefly in the summer or growing season. From the 1orst meridian to the Rockies the rainfall is from to to15 inches. Eastern Colorado is a region of small precipitation, with an average fall of about 12 inches and a maximum yearly fall rarely in excess of 20 inches. In western Kansas the precipitation of the driest year was 9.9 inches; of the wettest, 33.7 inches. The last-named illustration is typical of the wide differences between the extremes of rainfall in the arid and semi-arid portions of the West. The wettest years have a rainfall sufficiently great for agriculture; the means and minima are far below the necessary amount. Of the seven climatic and life provinces of North America, Plate I, but four fall within the limits of the United States, except that the cou Hen most or tropical province touches southern Florida and the lower valley of the Colorado, and the northernmost or Arctic province is developed on a few of the highest summits like Shasta in northern California and Blackfoot Mountain, Montana. The Boreal province is developed in the southern Appalachians of western North Carolina, eastern West Virginia, the Catskills, the Adi- rondacks, the White and the Green Mountains, and in the Superior Highlands of northern Michigan and Wisconsin and on all the main divisions of the Pacific Cordillera, where its upper limit coincides with the timber line. It is marked by a low mean annual temperature, generally between 32° and 4o°, by long, cold winters, and in general by a heavy snowfall. Tis forest growth is spruce and pine in New England, spruce and balsam in North Carolina, chiefly white pine in Michigan and Wisconsin, and spruce, fir, and cedar in the Pacific Cordillera. The greater part of the mountain forests of the United States is found in the Transition province which includes the cool temperate portions of the country with a generally high mean annual precipitation. The mean annual temperature is about 45°, but the temperature is in general marked by frequent and sudden changes. Snow falls through- out the entire province, though it is variable in amount owing to differ- ences of elevation, exposure, etc. In the northern portion of this province both broad-leaved deciduous trees and conifers grow; similarly in the west, scattered growths of oak, pifion pine, and sycamore of the lower mountain slopes mingle or shade into the spruce and yellow and white pines of the upper slopes. The province as a whole has few distinctive plants; it is marked rather by the mingling of southern species that here find their northern (on the mountains their upper) limit and of northern species which find their southern (or lower) limit of occurrence. PLATE | Scale of Miles 10 7% 50 2 0 , 10 200 300 Plate I. — Climatic and Life Provinces of North America. PHYSIOGRAPHIC, CLIMATIC, AND FOREST REGIONS 123 ! The western division of the Upper Austral province is so dry that agriculture without irrigation is impossible and tree growth is limited to the heads of the better watered alluvial fans and the banks of streams. The eastern portion of the province was originally covered with a varied and dense forest of hickory, maple, oak, and chestnut, as in the Ohio Valley and the Piedmont and the Appalachian Plateaus. The climate is warm temperate, with a long summer season. The Lower Austral province resembles the Upper in the western part of the country in its general treelessness except along the streams. East of the tooth meridian the climate is wetter, and throughout the entire Gulf and Atlantic Coastal Plain the temperature and rainfall are favorable to the growth of great forests of southern species of pine and of cypress. The winters are very mild, snowfall is absent, and the long, hot summers have an abundant rainfall. The mean winter tempera- ture is 40° to 52°, the summer temperature from 75° to 80°. FOREST REGIONS A single unbroken forest belt extends across North America, the spruce forest of Canada. Its northern border, a timber line determined by cold and physiological dryness, extends from Hudson Bay north- westward to the head of the Mackenzie delta, thence westward and southwestward across Alaska. Its southern margin is the 6oth parallel in the Canadian Northwest and the soth parallel in the Great Lake region. Black and white spruces, poplar, canoe birch, aspen, and tama- rack are typical growths; the presence of only a few species of trees is characteristic. The spruce forest includes a large part of the lake region of North America with an abundance of lakes and swamps (p. 565). The spruces and the gray pine grow on the uplands between lakes and swamps; poplar, dwarf birch, willow, and alder occupy the cold wet bot- tom lands. While the trees attain fair size on the southern portions of the belt in which they occur, they are never large, and decrease notably in size toward the north, where they finally become so stunted as to be of little economic importance. Southward from the broad transcontinental forest belt are an Atlan- tic forest, a Pacific forest, and a Rocky Mountain forest. The two intervening belts of country — the Great Plains and the Great Basin —are forestless though not treeless. This distribution is controlled largely by rainfall, though the distribution of species within each region is also controlled by insolation, temperature, wind velocity, water supply, and geographic relation to postglacial centers of dispersal. By the same token the forests are not distributed evenly over a given region, RAPHY x I FOREST PHYSIOG 124 (104 Jo ing “Way jo ydeq ‘SQ wor dep) “4saA\ 9Y} Jo suolziod papeysun oy} Ur ‘19}vAM\ Jo AOuaTOyap e Aq (Z) ‘eUISINO'T ULYJNOS UT UMOYS SI 19}VM Jo ssadx9 uv Aq (I) JSAIOJ 9} JO UOISN]OXd oY} /SUIO}JOq IOATI OY} SuoTe preMysam yof{ord suotsue}xe oy [-193Uy Q19YM VIIL SpoOOMpILY [vI]UID 9Y} JO Japsioq UII}sIM oy} UO JUsIvdde JsoW oI J9zVA\ PUNOIT Jo spoyo oY} ‘svxaJ, UoYINOS puv VpPIIO].J UIOYINOS Jo 4ySoIOF Jeordos} oy} UL UAOYS SI sapNqr] MOT Ur samnzesoduia, YSty Jo syoya oy} furaysAs uvrporjeddy ay} Suoye Yyos spuszxe yey} ysa10J ULOYJIOU JO WAL FuoT ay} Ul UMOYS SI s}saroJ puv sonqzesodur9} UO UOTwA]d Jo yoyo oy} ‘oJa ‘sasuvI sooag-suvIy, ay} ‘vIOYL YIMOS jo S| WLI oy} “Uosa1Q Jo sureyunoyy AN] 2Y} UL UMOYS SI S}SoIOJ PUL ][LJUIeI UO UOTZWAI]I JO JOTJa oY} ‘4SO\\ PUL JSeAY JO S]SVIUOD 4sSodIOJ 9Y} UL UMOYS ATIvIO Iv ][VJUIeI Jo spaya oy] — “fe “BT yP SWIBIITS 94} Suojy : - 1sayu04a-> ydooxg ‘ssojaai| oie soy papeysur aul ——~- awold + a (; 2S3uO4 UY Yj 2/, SaLVLS GF LENA Ein oe ne) SNOIDSY LSIYUO4S u-@ AN yusw i vitoxuva PHYSIOGRAPHIC, CLIMATIC, AND FOREST REGIONS 125 but vary from windward to leeward slopes, from warm southern to cool northern slopes, and from dry to wet situations as controlled by more local conditions. The eastern or Atlantic forest tract consists largely of hardwoods, though large tracts of conifers are also found; the western forests are principally conifers of unrivaled size and beauty and hardwoods are comparatively rare. The great variety and in some cases the great size of the trees of the Atlantic forest are distinctive features. The primary divisions of the Atlantic forest belt are (1) a belt of conifers, (2) a white pine belt, and (3) a belt of southern pines. The belt of conifers in which the white pine is the most important species extends from southeastern Canada and Massachusetts west- ward to northern Michigan, Wisconsin, and Minnesota. It reaches its best development in the light, dry, sandy soil of the glacial drift in the northern part of the southern peninsula of Michigan. White cedar, hemlock, fir, larch, and spruce are other conifers of minor importance in the tract. South of the white pine belt is a belt of hardwoods which extends through the Great Lake states and the larger part of the great Appa- lachian region. The most notable types of trees are many species of oaks, several kinds of hickory, the chestnut, basswood, magnolia, tulip tree, and cottonwood. In this belt the hickories reach their greatest size in the Ozark region, the oaks in the central and southern portions of the eastern United States, the tulip tree in Kentucky. Among the excep- tions to the characteristic growths of this division may be mentioned the spruce and hemlock forests on the summits of the Pisgah and other ranges in western North Carolina, where boreal conditions prevail. The Atlantic and Gulf Coastal Plain is occupied by longleaf, short- leaf, loblolly, and slash pines. The first two occupy dry sandy uplands for the most part; the loblolly pine grows best on the drier portions of the moist lowlands of eastern Texas. Among the trees of the Pacific coast forests in the United States the red or Douglas fir is the most important. It has its best development in the wet Puget Sound region, where it grows to a height of several hundred feet. It is associated with the tide-land spruce, hemlock, and red cedar. This portion of the Pacific forest is one of the densest and commercially one of the most valuable in the country. Southward and eastward the forest changes in character. In northern California, where the rainfall is heavy, the redwood is the most important forest tree, and between it and the fir forest on the north is a tract occupied by the Port Orford cedar. The dry and nearly treeless Great Valley of California divides 126 FOREST PHYSIOGRAPHY the Pacific forest into an eastern and a western section. The well- watered Sierra Nevada in the east has a heavy forest growth of sugar pine, red fir, yellow pine, and hemlock besides the famous Sequoia gigantea. The dry Columbia Plateaus have a discontinuous forest in which pine and larch are the most important types. The Great Basin region is still drier and supports an even more limited growth of pine and juniper at the lower elevations, and a scanty aonren of fir and spruce at the higher elevations. The Rocky Mountains forest extends from the borders of the sur- rounding plains and the intermont basins up to elevations of go0o to 11,000 feet and is much broken into forest islands by the restricted areas of mountain land which are sufficiently high to provoke an ade- quate rainfall from the prevailing westerly winds. Spruce grows luxu- riantly at elevations ranging from 8000 to 10,000 feet, and at lower altitudes yellow pine, red fir, and white fir are abundant. This growth is characteristic of the Rockies as far south as New Mexico, where the lower elevations of the mountains of the Trans-Pecos region cause the forest to disappear or to become restricted to a few higher ranges such as the Sacramento and Davis Mountains of western Texas and eastern New Mexico. Two small forest tracts of exceptional character deserve mention in even this brief description. In Florida the subtropical climate has given rise to an Antillean type of flora among which mahogany, royal palm, and mangroves are of chief interest. The second tract is in southern- most Texas, where vegetation occurs whose affinities are with the Mexican flora. CHAPTER X COAST RANGES Tue relief of the relatively dry western half of the United States has a high importance in the distribution of the forests because of the effects of relief upon two of the controlling factors of forest growth — temperature and rainfall. We shall therefore begin our consideration of the physiography of the United States by a study of the West. Of the 190 million acres of national forests, more than 185 million occur west of the eastern front of the Rockies, a fact that further increases the forester’s interest in the topography, drainage, soils, and rainfall of this vast region. SUBDIVISIONS The Coast Range system of mountains in the United States extends from southern California to the Straits of Juan de Fuca. The moun- tains of the system do not have uniform topographic qualities through- out but consist of four somewhat dissimilar sections. The southern section extends from southern California to the goth parallel; farther north are the Klamath Mountains, which extend from the goth to the 43d parallel; the third section embraces the low Coast Ranges of western and northwestern Oregon; the fourth includes the Olympic Mountains, which rise to heights of over 8000 feet and are the highest mountains of the system next to those in southern California. Coast RANGES OF CALIFORNIA The Coast Ranges of California terminate on the northern margin of Humboldt County (p. 141); beyond them to the northeast lie the Klamath Mountains, which are more closely allied to the Sierra Nevada Mountains than to the Coast Ranges in rock character and geologic history though not in geographic position. The Coast Ranges of California are sometimes regarded as ending on the south in Santa Barbara County, there giving way to the mountains of southern California. We shall here include the mountains of southern California with those of the coast of California to the 4oth parallel in a single coast group because of (1) the extension of the tectonic lines of the northern mountains into 127 128 FOREST PHYSIOGRAPHY the mountains of southern California and (2) the closely related fact that the movements along these lines — movements to which the larger topo- graphic features are due— date in both cases from the close of the Tertiary. A unity of both structural and topographic characters is thus given the entire group of ranges. The group is divided, however (largely on the basis of trend), into three subgroups: (1) the Coast Ranges proper; (2) a broad chain extending from Santa Barbara County to the eastern and southeastern side of the Colorado desert, with a general trend west- northwest, and including the San Rafael, Santa Ynez, Santa Susannah, Santa Monica, and San Gabriel ranges, a chain parts of which are known locally as the Sierra Madre, though the application is neither uni- form nor clear; and (3) the mountainous country of the Valley of southern California. The principal ranges of this group are the northwestward- trending Santa Ana and San Jacinto mountains, sometimes called the Peninsular chain.1 On both the north and the south the Coast Ranges are from 5000 to 8000 feet high; the elevations of the central portions are from 3000 to 4000 feet. San Lucia Peak, the highest peak of the central Coast Ranges, is less than 6000 feet high. In general the crests range from 2000 to 4000 feet. The eastern margin of the Coast Ranges of California rises abruptly from the floor of the great central valley of California as a well-marked continuous mountain front. At its southern end it is a dissected fault scarp; elsewhere a smaller amount of faulting has taken place, but every- where the eastern border represents a line of strong deformation. In a broad view the western margin of the Coast Ranges is not at the shore line but at the edge of the continental platform on the 600-foot submarine contour, where the sea bottom changes its slope abruptly from a previously gentle incline and plunges steeply down to depths of Sooo feet and more.” At the foot of this steep decline the sea bottom again assumes low gradients. The slope constitutes a notable moun- tain front rising from the floor of the Pacific and forming the natural western boundary of the coast system of mountains. It is interpreted as a great submarine fault scarp or series of fault scarps comparable to those that form the eastern front not only of the Coast Ranges of Cali- fornia but also of the Sierra Nevada. At the base of the steep sub- marine scarp, dredgings (Tuscarora explorations) at the depth of 12,000 feet have brought up fragments of bituminous shale which are con- 1 A. C. Lawson and others, Section on Geology, The California Earthquake of April 18, 1906, Carnegie Inst., vol. 1, pt. 1, pp. 2, 3 et al. 2 Andree’s Handatlas, bathymetric chart, No. 157. COAST RANGES 129 116° GEOMORPHIC MAP OF CALIFORNIA SHOWING THE DIASTROPHIC CHARACTER OF THE RELIEF AND THE MOST IMPORTANT KNOWN FAULTS Seale of Miles 50 100 Monterey Bay (eS) STA. CRUZ \2 ¢ Ds a Sexe) 4 SS s Angel STA. ROSA Los Angeles CATALINA SAN YN San Diego Sb CLEMENTE 120° Fig. 24. —Map of California. The heavy lines indicate the principal faults. 130 FOREST PHYSIOGRAPHY sidered to be talus débris of so recent origin as not yet to have been buried by oceanic sediments. The coastal scarp above sea level is not everywhere regular in de- velopment; the most noticeable interruption is the Bay of ‘Monterey and adjacent slopes, which form parts of a synclinal trough whose axis is at right angles to the trend of the coast and of the Coast Ranges as a belt. A second interruption of the continuity of the coastal scarp is at the Golden Gate and is due to a depression of the Coast Ranges which resulted in the drowning of the lower portions of land valleys that formerly crossed the coastal mountains. The Point Rees penin- sula is a third important break in the continuity of the coast line of California and is due to the manner in which the depression east of the ridge forming the peninsula has been drowned; the northern end of the valley is occupied by Tomales Bay, the southern by Bolinas lagoon.t The Coast Ranges of California consist on the whole of a series of parallel ridges composed of sedimentary strata (Cretaceous? and later) that have been deformed on broad lines by deep-seated causes. There has been crumpling of the strata besides a certain amount of igneous eruption, but the major features are due to the effects of great dissection and later block faulting on a large scale, attended and fol- lowed by erosion. The character of the relief is in many cases markedly diastrophic and the relief features commonly have the rectilinear qual- ity associated with pronounced faulting, which also explains to a large degree the parallelism of the ridges. Examples are Castle Rock Ridge, Cavilan range, the Santa Cruz range, and many others whose borders are marked by the San Andreas fault, the Castle Rock fault, etc., which during the California earthquake of 1906 were the loci of maximum earthquake intensity.® The most important line of faulting in the Coast Ranges of California is the Rift, as it has been termed, or the San Andreas Rift, a name taken from the San Andreas valley of the peninsula of San Francisco. The Rift is a continuous topographic depression for at least 190 miles from Point Arena to San Juan, and in this part of its course is nearly straight, following an old line of seismic disturbance which has a much greater extent — that is to say, from southwest of the Point to southern California, or about 600 to 7oo miles. Indeed the Rift may extend much farther to the south and may be associated with the origin of the 1 Lawson, loc. cit., pp. 12-15. 2 For geologic time names consult Appendix D. 3 Atlas of maps and seismographs accompanying the Report of the State Earthquake In- vestigation Commission upon the California Earthquake of April 8, 1906, Maps 1, 22, and 23. Also the Santa Cruz quadrangle, U. S. Geol. Surv. COAST RANGES 131 Colorado desert and the Gulf of California. The physical habit of the Rift valley, for example in the Bolinas-Tomales section, is that of a remarkably straight depression, with the southwestern wall steep, the northeastern wall gentle. The character of a pronounced topographic depression, however, is not everywhere sustained. The southern end of the great Rift may be traced for an unknown distance along the base of the mountains bordering the Salton Basin upon the northeast, where it probably dies out gradually. It is coin- cident with long and narrow valleys whose orientation is controlled by faulting along the Rift but whose detailed features are in large measure determined by erosion upon the exposed edges of formations of varying hardness. The depressions which constitute the major Rift along the southern margin of the Mohave desert appear to be almost wholly diastrophic. The steep northern flank of the San Rafael and San Gabriel ranges on the south side of the Mohave desert are degraded fault scarps, the walls of the great Rift valley. The exact share in all these vari- ous sections of the Rift valley that may be ascribed on the one hand to crustal deformation of the fault block type and on the other to erosion has not been determined. For miles at a stretch the earth on one side or the other of the fault in the southern part of the Coast Ranges has sunk in such manner as to give rise to basins and cliffs measured in terms of several hundred feet. The individual ridges of the Coast Ranges of California have a pro- nounced parallelism in a direction somewhat oblique to the main trend of the coast, so that they tend constantly to emerge upon the coast in the form of northwestward-trending peninsulas. The courses of the longitudinal valleys correspond either to the strike of the rocks or the trend of the fault lines and are oblique to the general trend of the coast range belt. The general drainage is therefore termed subsequent, for the streams have extended themselves along belts of weak rock or along fault depressions at the expense of an earlier drainage crossing the region in a westerly direction or transverse to the structure. Short sec- tions of the streams cross the ridges in steep-sided valleys or gorges, and these only may be termed antecedent. The tops of the ridges in some respects are more or less flat and pre- sent the character of a rolling, mature upland; but more commonly they are determined by the intersection of the slopes of adjacent valleys; even in the latter case, however, it is generally true that the ridge crests over wide areas reach about the same altitude and in a broad view give the impression of an upland with fairly uniform elevations and gentle '1 Lawson, loc. cit., p. 20. 132 FOREST PHYSIOGRAPHY slopes. The stream valleys, cut below the level of the dissected upland, are usually wide-bottomed in the softer and narrow-bottomed in the harder rocks.1 The Santa Lucia Range illustrates many of the general features of the region. It is the dominant mountain range of the coast of Cali- fornia for over 100 miles between latitude 35° and 36°30’ N., Fig. 24. For much of this distance it rises boldly from the Pacific Ocean and forms the most picturesque portion of the California coast. In places the spurs of the range terminate in cliffs several hundred feet high; in other places the range is bordered on the seaward side by a gently sloping platform or terrace which is from 4o to 80 feet high on its cliffed outer margin and roo feet high on itsinner margin. This platform is primarily a wave-cut terrace, though its surface is thinly covered with wash from the. bordering hills.2, The Santa Lucia Range has an even sky line many miles long, and a summit from 2 to 4 miles wide. Its regular front is a bold, compound, fault scarp. The range is traversed by narrow canyons which open out headward into broad valleys in an advanced stage of topographic development. In this respect the range resembles many others among the Pacific mountains. An earlier surface, in some places softened and subdued with moderate waste-covered slopes, in other places a true peneplain, was deformed by faulting. The summit levels of the uplifted fault-blocks (the present ranges) display remnants of the ancient smoothly-contoured surface in strong contrast to the steep borders of the ranges sharply outlined by more recent faulting and now in process of vigorous dissection. Large portions of the Coast Ranges are unknown even through re- connaissance surveys. Among the known portions the Santa Cruz section between San Francisco Bay and the Bay of Monterey presents features of special interest. Here the parallelism of the valleys and ridges is apparent in the larger features of the topography but is less marked or absent in the minor relief. The main ridges have a steep northeast slope bordered by a series of valleys lying along the San Andreas Rift. The lines of the major folds of the Santa Cruz region are marked by more or less continuous valleys, and in the case of both these larger valleys and the main ridges the topographic and geologic features are in sympathetic relation.* The hillsides of the region are generally covered with a deep coating of soil, and cliffs are rare, owing both to the friability of most of the rocks and to the advanced state 1 Lawson, loc. cit., p. 20. 2 H. W. Fairbanks, San Luis Folio U. S. Geol. Surv. No. tor, 1904, p. 1. 3 Branner, Newsom, and Arnold, Santa Cruz Folio U. S. Geol. Surv. No. 163, 1906, p. 1. COAST RANGES 133 of topographic development which was reached before the last and recent uplift. An unusual feature of the topography of the Santa Cruz region is the occurrence of very steep yet soil-covered hillsides; 35° to 40° slopes are not uncommon, and in one place is found a soil- and vegeta- tion-covered hillside with a slope of 50° from the horizontal. There is a dense growth of timber and underbrush over much of the area, which does not prevent the thick covering of soil from being frequently in- volved in landslides in the belts of greatest faulting and folding.! The Coast Ranges of northern California include, besides the moun- tains proper, a coastal tract which was eroded (Pliocene) to the form of a peneplain. The coastal peneplain was then uplifted and its streams intrenched; it now forms a dissected plateau with long and roughly level-topped ridges separated by equally long, narrow valleys. The ridges are remarkably constant in general altitude, and the sky line is essen- tially level. In a general perspective the view is that of a plain or sloping plateau of low relief. The peneplain was uplifted to an ele- vation of 1600 feet above the sea on the seaward margin, and to 2100 feet on the inner margin. The mountainous tract adjacent on the east participated in the same movement. In Humboldt County several sharp peaks rise abruptly above the general level of the dissected plateau to 4000 or 5000 feet, but they are clearly encircled by remnants of the plateau which give to the mid-slopes of the peaks a distinctly terraced aspect. The peneplain may be followed in among clusters of mountain peaks and ridges and extends at least as far as the Bear River ridge. That the present dissected plateau was once a peneplain -is inferred from the facts that the rocks composing it are of varying ages and of varying degrees of hardness, and that the general surface of the region bevels rather evenly across the deformed strata. On the summit of some of the ridges of the plateau numerous water-worn pebbles have been found, at 1600 feet, which are reasonably interpreted as remnants of larger bodies of stream gravels formed upon an erosion surface. The coastal peneplain grades into a region of stronger relief on the east where the stream courses were still completely under the control of geologic structure at the end of the first erosion cycle and flowed in mature subsequent valleys which were inherited by the streams of the second or present cycle of erosion. The abrupt coastal margin of the uplifted peneplain of northern California has given rise to a youthful 1 Branner, Newsom, and Arnold, Santa Cruz Folio U. S. Geol. Surv. No. 163, 1906, p. To. 2 A. C. Lawson, The Geomorphogeny of the Coast of Northern California, Univ. Cal. Bull., Dept. Geol., vol. 1, pp. 242-244. 134 FOREST PHYSIOGRAPHY topography along the coast; the coastal canyons are narrow and precipi- tous, and V-shaped profiles predominate. In the middle stretches of the streams degradation is less intense and the topography appears somewhat less rugged. | Recent events following the uplift and dissection of the coastal pene- plain of California are a subsidence of at least 370 feet at the mouth of the Sacramento River which flooded the lower portions of that valley and gave rise to the magnificent harbor of San Francisco. The drowned mouth of the river once discharging across the Coast Ranges at this point is known as the Golden Gate. The last episode in the region has been a slight uplift in the vicinity of the Straits of Carquinez.! Fig. 25. — Coastal terraces produced by wave erosion, west of Santa Cruz, California. (U.S. Geol. Surv.) The uplift of the coastal peneplain cf northern California was not accomplished in a single continuous movement but was interrupted by many halts. During these periods of relative stability there were formed well-developed ocean terraces which are among the most promi- nent features of the coastal topography. Such terraces were always involved in later uplifts and now stand at high levels, the highest representing the algebraic sum of all coastal changes whether of uplift or depression since the beginning of the last series of changes in the level of the land. The highest terrace of northern California is about 1 A. C. Lawson, The Geomorphogeny of the Coast of Northern California, Univ. Cal. Bull., Dept. Geol., vol. 1, pp. 270-271. COAST RANGES 135 1500 feet above sea level. Below this are prominent terraces at 1400, 1180, 760, 440, 350, and 280 feet, respectively, above sea level, with many less prominent terraces at intermediate levels. The lower terraces have all the characters associated with wave and current origin, such as a rather regular seaward slope, upturned strata smoothly planed off, residual stacks, beach bowlders, and sea cliffs with horizontal base lines. The higher ones are usually not so clear, though even the highest have sufficient definition in the form of sea cliff, sloping terrace, and bowlder beach to make its character certain." MOUNTAINS OF SOUTHERN CALIFORNIA In the southernmost division of the coastal mountains of California (see p. 127) faults have also played a very important part in the topog- raphy. Both the northern and southern sides of the San Gabriel range are determined by a profound fault; the range may be interpreted as a horst thrust up between two bounding faults. Since uplift the range has been thoroughly dissected and older surfaces of erosion destroyed. “One seeks in vain for horizontal lines along the San Gabriel tops; a confusion of peaks and ridges of discordant and seemingly unrelated heights makes up the mountain mass. . . . [They] present a labyrinth of canyons and ridges and peaks, with no level areas of any size. The ridges have narrow summits; the peaks are sharp; the streams are all evenly graded from source to mouth.” 2 The Santa Ana Mountains are a tilted, seaward-sloping mountain block with a very straight and abrupt fault scarp that faces the north- east and overlooks the Perris plain. The block is an elevated and as yet but little dissected peneplain (Cretaceous) with remnants of younger (Tertiary) deposits upon it, indicating that it has in part at least been resurrected in recent times from a buried condition. It is thought that the same tilted block structure extends beyond the Santa Ana Mountains southward to the international boundary and even beyond. Both sides of the San Jacinto Mountains are precipitous and probably determined by faults, so that the ridge has very bold margins. Among the drainage features of these mountains are the interesting valleys of the Santa Ana and Santa Margherita rivers which are antecedent to the tilting of the region; they persisted in their southwestward courses during the development of the fault scarps, and now cut squarely across the range, draining the valley lands on the northeast.’ 1 A. C. Lawson, The Geomorphogeny of the Coast of Northern California, Univ. Cal. Bull., Dept. Geol., vol. 1, pp. 246-247. 2 W.C. Mendenhall, Ground Waters and Irrigation Enterprises in the Foothill Belt, South- ern California, Water-Supply Paper U.S. Geol. Surv. No. 219, 1908, p. 17. 3 A, C. Lawson and others, The California Earthquake of April 18, 1906, Carnegie Inst., vol. 1, pt. I, pp. 23-24. 136 FOREST PHYSIOGRAPHY SAN BERNARDINO RANGE The San Bernardino range of southern California is a distinct topo- graphic unit and does not have a close genetic relationship with the other members of the Coast Range System in southern California. It is much younger than the San Gabriel range and appears to have had a history different from that of the San Jacinto range south of it. The relief of the mountains is outlined upon an uplifted fault block once of somewhat more regular development than at present. Rem- nants of an old surface of moderate relief, broad elevated valleys, ’ Fig. 26. — Redlands and San Bernardino and San Gorgonio Peaks, San Bernardino Mountains, Cali- fornia. (Mendenhall, U. S. Geol. Surv.) plateau-like ridges, and several interior basins like those in the Mohave desert on the north, are the principal secondary topographic elements. At its western end is displayed a long even sky line at elevations between 5000 and 6000 feet above the sea. (73 . . . there are many wide upland valleys, forested and grassy glades, and lakes or playas like Bear Lake and Baldwin Lake. Where these upland levels are attained it is difficult to realize that one is actually in the high mountains. The surrounding topographic forms are rounded and gentle, the level areas are extensive, the streams meander placidly through broad meadows, and the topographic type is that of a rolling country of moderate elevation. But as the edge of these interior uplands is approached the streams plunge into precipitous canyons, the slopes are as steep as earth and rock can stand, the roads and trails twist and turn and double to find a devious and precarious way to the valleys below.”’! 1 W. C. Mendenhall, Ground Waters and Irrigation Enterprises in the Foothill Belt, South- ern California, Water-Supply Paper U.S. Geol. Surv. No. 210, 1908, p. 17. °.Pickaninny Butte SVN; Vv E Sy Ly = head Canyon ING Ter nace yy) ZS Horsethie Arrastre SX _ Plates SB Saragoswi Springs Ua Zn, | Kittitas County, orrow County, Washington Washington Oregon Tnsolubl 4 i a nsolublesmatterss Cerbat Mts. Detrital-Sacramento Valley and other areas in black represent lava flows; the valleys are deeply filled with alluvium; the Cerbat Mountains are composed of sedimentary rocks. Fig. 67. — Geologic section from Colorado River to Colorado Plateaus (right). Geol. Surv.) 248 FOREST PHYSIOGRAPHY (x) Zone of cactus, yucca, agave, scanty grass, 3000 to 3500 feet. More luxuriant vegetation in the vicinity of water. (2) Zone of Obione and Artemisia (greasewood and sage-brush), poor grass, diminished growth of cactus, altitude 3500-4900 feet. (3) Zone of cedar (Juniperus occidentalis), few cactus, 4900 to 6800 feet: (4) Zone of pine and fir, 6800 to 10,800 feet. Scale of Miles feasgid Let ise Fig. 68. — Waste-bordered mountains of the Arizona Highlands, Camelsback quadrangle, U. S. Geol. Surv. Note the lack of a permanent stream, the large number of intermittent streams as shown by the dotted lines, and the ragged character of the mountain slopes. Contour interval, 250 feet. These zones are lower on eastern and northern than on southern and western slopes and the amount of canting increases with increasing elevation (p. 293). The quaking aspen is seen below 7500 feet, likewise the fern (Pteris aquilina). Above 7000 feet the white oak accompanies ARIZONA HIGHLANDS 249 the pine but is never found in great quantities, principally in small patches or groves. In some instances the pines occur in splendid forests. The character of the tree growth in the extreme southern part of the Arizona Highlands has been analysed by Mearns.? He found (1893) the Mexican white pine (Pinus strobiformis) at the summits of the main peaks of the San Luis Mountains south of the boundary line (7874 feet); also on the Animas peaks in New Mexico (8783 feet); and in the San José Mountains (8337 feet), where a few trees grow close to the summit of the main peak. It is a common tree on the highest peaks of the Huachuca Mountains where it occupies a considerable area and descends as low as 6550 feet. It is an interesting fact in the distribution of this tree that it belongs to the Canadian life zone and is usually asso- ciated with the Douglas spruce, aspen, etc. The yellow pine has a vertical range (on the Mexican line) from 6200 feet in the San José Mountains of Sonora to 8500 feet in the Huachuca Mountains of Arizona. The Douglas spruce is found on the San Luis, Animas, and Huachuca Mountains, on all of which it reaches the summits and ex- tends as low as 6000 feet, in cold wet ravines, and is not found on any other mountains of the boundary strip.’ On the Dog Mountains in New Mexico the regular juniper zone extends from 6000 feet up to the summits, but in most canyons it de- scends to the base of the mountains. There is a well-marked juniper zone at 7500 feet on the west side of the San Luis Mountains also. The desert yucca in the fifty-mile desert west of El Paso occurs in open forests spread over large areas where the largest trees grow to a height of 16 feet. The aspen or quaking asp has a vertical range on the Mexi- can line from 7690 feet in the San José Mountains of Sonora, Mexico, to 9472 feet in the Huachuca Mountains in Arizona. The Fremont cotton- wood (Populus fremontii) is the most common, beautiful, and valuable shade tree in the whole Mexican boundary region. It grows natu- rally on almost every stream along the boundary and is found planted around the houses and along the irrigation ditches of almost every ranch. In deep narrow canyons its stem is tall and slender, but in open spaces the isolated trees have full round tops with spreading and often drooping branches. The vertical range of the cottonwood is from sea level to 6100 feet in the Huachuca Mountains near the boundary.* 1 Report upon Geographical and Geological Explorations and Surveys West of the tooth Meridian (Wheeler Surveys), Geology, vol. 3, 1875, pp. 603-604. 2 E. A. Mearns, Mammals of the Mexican Boundary of the United States, etc., Bull. U. S. Nat. Mus. No. 56, pt. 1, 1907. 3 Idem, pp. 38, 39, 40. 4 Idem, pp. 43-48. 250 FOREST PHYSIOGRAPHY A common tree or shrub through the desert Southwest is the mes- quite (Prosopis glandulosa). The vertical range is from sea level and even below sea level in the Colorado desert up to about 5500 feet. In the deserts of New Mexico, Arizona, and California it is a shrub which obstructs drifting sand, thus forming mounds of sand and lines of sand hills; in the most fertile places along the Colorado River and its tribu- taries it is a tree of considerable size. Along the Santa Cruz River in Sonora are forests of unusually large mesquite, with some individuals 25 feet in diameter and 50 feet high.! The San Luis Mountains are composed largely of calcareous rock and are steep and rough. Where a soil covering has been formed they are wooded from a well-marked dry timber line at 5250 feet to the sum- mit at 7870 feet. Below the lower timber line the country is covered with grass and in places with patches of mesquite and chaparral. The forest trees at the lower timber line of the San Luis Mountains are mostly evergreen-oak (Quercus emoryi), though in the low canyons grow cypress, walnut, cherry, sycamore, and gray oak (Quercus gresea).? The trees of the Animas Mountains (northward continuation of the San Luis Range) in New Mexico are the same as those of the San Luis, with the addition of a zone of quaking aspen. The lower timber line in the Animas mountains is at 5250 feet except where springs occur that support a belt of fine oak timber in the moist canyons far below the main timber line. In one instance a straggling line of oaks is ac- tually continuous across the valley between the Animas and San Luis Mountains, joining the two timber lines of these mountains down two long canyons.* REGIONAL ILLUSTRATIONS CLIFTON DISTRICT In portions of the Arizona Highlands, as for example in the vicinity of Clifton, near the eastern border of the state, the topography be- comes far less regular than is generally the case. Looking north of Clifton it is impossible to discern any well-defined mountain system. The whole region north of Gila Valley at this point appears as a maze of short ridges, small plateaus, and insignificant peaks. The topographic complexity of the highlands in the Clifton district is explained by the geologic structure. A core of older rocks (granites, limestones, and sandstones) was deeply and irregularly eroded, and at 1 E. A. Mearns, Mammals of the Mexican Boundary of the United States, etc., Bull. U. S. Nat. Mus. No. 56, pt. 1, 1907, pp. 59-60. 2 Tdem, pp. 89-90. 3 Idem, p. 92. ARIZONA HIGHLANDS 251 a later time (Tertiary) was partially covered by great masses of volcanic flows (rhyolites and basalts), with great variations in thickness and char- acter. The drainage developed upon the lavas after their extrusion. was consequent upon the lava flows; those portions of the region not affected by lava flows preserved their original drainage. The conse- quence was that extremely irregular drainage courses were made still more irregular by the differences in hardness between flat-lying basalt and deformed rock of older age.1 The only regular features of the Clifton district are the small plateaus due to volcanic accumulations or to the regular and broad uplift of sedimentary rock; but even these plateaus are but dimly discerned and all of them are deeply dissected by canyons and furrowed by a maze of shallow and wide-spreading ravines. The arboreal vegetation of the Clifton district is found at elevations above 6000 feet, though below this elevation the ridges generally sup- port a certain amount of agave, yucca, and low cactus. Above the 6000- foot level stunted juniper and cedar are quite common and are used as firewood; on the higher slopes a growth of manzanita bushes and stunted oak is also found. The heaviest timber grows in the sheltered mountain basins at altitudes of 5000 to 6000 feet and consists largely of yellow pine. Along some of the river bottoms are large groves of cottonwood. Many of the dry mountain spurs are covered with pion and juniper.” BRADSHAW MOUNTAINS The higher peaks of the Bradshaw Mountains district are composed of gneiss, granite, and schist, the granite having been intruded into the schists. Differential erosion has resulted in a high relief, due to the more resistant character of the intrusive granite. In some places quartzite.combs in the granite stand prominently above the general level of the wide valleys formed upon the less resistant schist and may be traced for miles by their distinctive and bold relief. Toward the northwest the Arizona Highlands are also marked by a large number of rather extensive lava flows which have covered over the older topography and simplified the contour of the surface. Among them is Black Mesa, 10 miles east of the Bradshaw Mountains, Bigbug Mesa, 15 miles northwest of Black Mesa, and many others of lesser extent. They are striking forms, for they lie in a region of generally rugged topography whose irregularities have been devel- 1 Waldemar and Lindgren, Clifton Folio U. S. Geol. Surv. No. 129, 1005, p. I. 2 Idem, p. 2. 252 FOREST PHYSIOGRAPHY oped upon rocks of complicated structure and variable hardness. They are almost without soil, since they consist of durable basalt recently formed. The basalt weathers into spheroidal fragments of variable size which cover the surface and make the so-called “malpais”’ of the region. Between the fragments finer waste accumulates in small quantities and supports a thin growth of grass in the rainy season. The schists of the district weather very slowly and their soils are thin, so that the outcrops of the steeply inclined or vertical strata are visible Fig. 69. — Bradshaw Mountains, looking northwest near Goddard. The mountains are composed of granite; the plain in foreground is underlain by basaltic agglomerate. (U.S. Geol. Surv.) for miles. The quartz-diorite and granite weather more rapidly and are covered with a sandy soil which supports a good forest growth in the higher mountains of the region toward the north. The quartz-diorite weathers most easily of all and is noted for the characteristic basin or park-like forms developed upon its outcrops. The semi-aridity of the plains and valleys of this district has resulted in the development of a characteristic arid vegetation. The common vegetation of the lower slopes consists of cactus, yucca, maguey, paloverde, “cat claw,” etc. In favored localities there are stunted growths of oak and manzanita, and in the arroyos one may find larger oaks and sycamores in some quantities. On the mountains where greater rainfall is precipitated on account of the greater elevation pine and fir find a congenial habitat. ARIZONA HIGHLANDS 253 In the Bradshaw Mountains the precipitation is greater than on the surrounding plains. Heavy thunder showers occur almost daily in July and August, and during the winter months the mountains are fre- quently covered with snow. The heaviest timber grows in the moun- tain basins at 5000 and 6000 feet and is largely yellow pine and its varie- ties. Some of the river bottoms have by contrast thickets of willow, mesquite, and alder, and groves of cottonwood, while the mountain spurs are frequently covered with thick mats of shrubs or dense stands of pin oak, nut pine, greasewood, and juniper." SANTA CATALINA MOUNTAINS The Santa Catalina Mountains ten or twelve miles north of Tucson are among the most impressive ranges of the province and indeed of the Southwest. Their deeply dissected, bold, picturesque slopes rise to a series of exceptionally ragged peaks. Perpendicular cliffs and sharp ridges are common, among them ‘“‘ The Needles,” a series of long slender precipitous points which crown the summit of a sharp granite ridge that rises 3000 feet above the plain at the mountain base. The highest point in the range is Mount Lemmon, which rises to 10,000 feet, or 7000 feet above the plains. The ranges composing these mountains are sub- parallel, extend nearly east-west, and the whole belt is about 50 miles long and almost as wide. The southern slope is especially rugged, so that even the cacti, hardy yuccas, and Spanish daggers have a hard struggle to maintain themselves on the barren rocks. Oaks and juniper are occasionally found in some sheltered alcove, and the summits of the higher mountains and large portions of the northern slopes are covered with pine and fir.” EASTERN BORDER FEATURES The northeastern edge of the Arizona Highlands is formed by the Grand Wash cliffs and their continuation — the Yampai Cliffs — which rise in a precipitous manner 4000 feet or more above the plains that on the north stretch westward from their base, Fig. 67. The continuity of the westward-facing escarpment is broken by several canyons. Here in a few places the confluent alluvial fans constitute a piedmont foreland of slight development. The lower portion of the cliffs is steep but not even approximately perpendicular. It is composed of granite, while the upper portion of the slope is of limestone and decidedly precipitous. In the northern portion of the Arizona Highlands and in the Sacra- 1 Jaggar and Palache, Bradshaw Mountains Folio U.S. Geol. Surv. No. 126, 1905, p. 1. 2 J. W. Toumey, La Ventana, Appalachia, vol. 8, 1897, pp. 225-232. 254 FOREST PHYSIOGRAPHY mento Valley region in northwestern Arizona the mountain escarp- ments face the plateau and are due to faulting. Specific localities are Globe, Arizona, and the mouth of the Grand Canyon. Farther from the edge of the plateau sedimentary rocks do not occur in many cases, and it is difficult to tell whether the mountains, composed of crystalline rock, are the products of local uplift or of circumdenudation. In general the mountains parallel the bordering plateau and become smaller and more isolated the greater their distance from the plateau. As the mountains grow less conspicuous the valleys broaden to greater and greater width and finally blend into each other in such a manner as to form a plain completely surrounding isolated mountain groups. The topographic character of this portion of Arizona lies between two extremes of well- defined fault-block mountains and narrow gorge-like valleys near the Colorado Plateaus and broad alluvial plains surrounding single mountain groups at a distance from the plateau. Truxton Plateau serves as a type of the eastern border features. It is comparatively level and extends from the Yampai cliffs on the east to the Cottonwood and Aquarius cliffs on the west. It lies about 5000 feet above the sea and consists of denuded granite whose depressions have been almost filled with eruptive rock but whose higher portions project above the lava. Truxton Plateau is described as a lava- covered peneplain which has been slightly dissected by a few streams that have cut narrow canyons.! The recent uplift of Truxton Plateau is indicated by the rapid deepening of the stream valleys as they approach the cliffs to the west, while their courses within the plateau are distinctly shallow.? For general remarks on Soil Investigations and the Soils of New Mexico and Arizona see G. K. Gilbert, U. S. Geol. Surv. West of the tooth Meridian (Wheeler Surveys), vol. 3, 1875, Pp. 594-507. For climatic conditions, idem, pp. 598 et seq. Also idem, pp. 603 et seq., for the geographical distribution of plants in this region. RAINFALL AND RINGS OF GROWTH The conditions of growth of the forests of yellow pine in the northern Arizona region have suggested that they might form climatic registers of great importance and that a study of the rings of growth may indicate the character of the climate during the life of the individual tree. This matter has recently been studied with some very interesting results.* 1 W. T. Lee, A Geologic Reconnaissance of a Part of Western Arizona, Bull. U.S. Geol. Surv. No. 352, 1900, p. 21. 2 For a brief discussion of the physiography of Arizona and the topography of the Globe quadrangle with some excellent cross sections see F. L. Ransome, The Geology of the Globe Copper District, Arizona, Prof. Paper U. S. Geol. Surv. No. 12, 1903. 3 A. E. Douglass, Weather Cycles in the Growth of Big Trees, Weather Rev., June, 1900, PP. 225-237. ARIZONA HIGHLANDS 255 It has been shown that at the 7000-foot elevation at which these trees grow the seasons are very sharply defined; the mean temperature for January is 29°, that for July 65°. Consequently there is a sharply seasonal character to the tree growth. A narrow red ring is formed during the autumn and winter and a broad, soft, white ring during the summer. Under the microscope the winter cells look lean and emaciated; the summer cells are round and well fed. The winter ring is thin, hard, and pitchy; the summer ring is wide, white, and pulpy. It appears probable that the red winter rings are governed directly by the low temperature and the white summer rings by the abundance of moisture. About twenty sections have been measured by micrometer scale and the result is thousands of readings covering a period of from two to five centuries. On theoretical grounds it would seem that these rings of growth ought to register the rainfall, for they are a measure of the food supply, which depends entirely upon moisture, especially where the supply is limited and the life struggle of the tree is against drought and not against its fellows or members of other species. The measurements show that the rainfall curves and curves of growth have a really remark- able resemblance. An analysis of the curves of growth for longer periods during the life of the tree indicates that there is a direct con- nection between curves of growth and curves of rainfall in 21.2 and 32.8 year periods! with suggestions of shorter periods, especially in the case of the r1-year period already determined by Bigelow.’ 1 —. Briickner (Vienna), Klimaschwankungen seit 1700, nebst Bemerkungen iiber die Klimaschwankungen der Diluvialzeit, 1890. In this paper the author assembles the data for the changing level of the Caspian Sea and its tributaries to show a strikingly regular rise and fall in cycles of about 35 years. Further investigations by Briickner show that these oscillations hold for much larger areas than the Caspian region, probably for the whole world. Still later analyses of rainfall curves in numerous localities have brought out the generality of this fact and the wide application of the law of climatic oscillations. 2 F. H. Bigelow, Studies of the Diurnal Periods in the Lower Strata of the Atmosphere, Weather Rev., July, 100s. CHAPTER XVII COLORADO PLATEAUS THE physiographic province known as the Colorado Plateaus is roughly circular in shape and embraces portions of the states of Utah, Arizona, Colorado, and New Mexico. The Grand Wash Cliffs on the west, the Uinta Mountains on the north, the Colorado ranges and the Trans- Pecos Highlands on the east, and the Arizona Highlands on the south are the most conspicuous border features. The boundaries are definite on the west, north, and south, but the eastern boundary is in many places indefinite, partly because of its topographic character, partly be- cause it is the least-known portion. The outline displayed on the gen- eral physiographic map, Plate IT, follows rather closely that assigned by Dutton.! Its most prominent and general characteristics are (1) the flat and but partially dissected upper surfaces of the various members of the province; (2) the distinct boundaries of the individual members as determined by (a) strongly developed, northward-trending fault scarps and (b) eastward-trending lines of cliffs developed most strikingly upon the out-cropping strata of the High Plateaus of Utah; and (3) the great canyons that ramify through almost every large section but are thoroughly developed in the central portion. The most striking physiographic features of the province are the Colorado River and its world-famous canyon. An early view of the course of the river ascribed its origin to a great fault which the river followed and modified. The detailed studies of later years show that certain portions of it are demonstrably located upon faults but that the course as a whole is independent of faulting; and the facts as at present known point to a consequent origin upon either (1) the deformed surface of a peneplain or (2) the present structural surface when it was in a different attitude.’ While the surfaces of the individual members of the Colorado Plateaus are smooth or gently undulating, level or slightly tilted, and so undis- 1 C. E. Dutton, Mount Taylor and the Zuni Plateau, 6th Ann. Rept. U. S. Geol. Surv., 1884-1885, pp. 114-117, and Plates 11 and 12. 2 H. H. Robinson, The Tertiary Peneplain of the Plateau District and the Adiicent Coun- try in Arizona and New Mexico, Am. Jour. Sci., vol. 24, 1907, p. 120. 256 _—-—. eee) a SM (oper elena = ae —-—- —__- —_. ———-— -_ — . — ce 4s, ———— TS D SN POATES, ENG. CO.| LeL. 114 saa = ty eae han eel Sa oe: ie? | | 4 PLATE II 113 MAP OF THE COLORADO PLATEAUS SCALE OF MILES ——= 25 50 v0) Beckhwi; Plat. ua y “ SS ‘ \ s |, Y Bear Ri fig Dolore HES me Mt. > ‘Slaglathay | A TEW MEXICO Needle Ws, 6 wes Yj OF, Plateau ot Hurricane are: [marsh Pass - x “tg, ~ ee le Jini Mesq~ Chin Lee Valley de Chelly i ZilhtTusayan (qs feholskal yo Pachado t i 1 L PS ie die fh GaP As a 6 Ss Mts. fe. ake Gallup ™ec_ We n ray *Zuni= $ “ @ Mts. Fring. cA peter Holbrook SWART ANU 3h ° LLIPOATES, ENG. CO., N.Y. lit 113 Longitude 112 Plate II. — Colorado Plateaus. sected as a whole that the level sky line, the broad expanse, the great extent of struc- tural surface are among the most important elements of the relief, yet it must not be thought that the region can be described even in general phrase as level. The great depth to which the canyons have been cut and the great breadth to which they have been widened by the action of sec- ondary erosional pro- cesses have resulted in strong relief so that the province is one of the ruggedest in the Wests Unlike “the mountain sections of the West its relief is not to any important degree the product of upward departures from the general level of the region but of down- ward departures. In- stead of mountain flanks we have here prodigious canyon walls diversified by innumer- able cliffs; instead of lofty ridges and peaks we have here profound chasms. Exceptions to this rule are the vol- canic and_ laccolithic COLORADO PLATEAUS 257 A group of laccolithic mountains breaking the evenness of Also called El Late Mountains (Hayden Surveys). Looking from the same point as in Fig. 71, but in a southwest direction, Tig. 70, — Ute Mountains, from Cortez, Colorado. (Photograph by H. Brigham, Jr.) the plateau surface. 258 FOREST PHYSIOGRAPHY masses that rise above the general level to a truly mountainous height, such as Mount Taylor (southeast), the San Francisco Mountains (San Francisco Plateau), Mount Dellenbaugh (Shiwits Plateau), Mounts Trumbull and Emma (Uinkaret Plateau), and the Ute Mountains of southwestern Colorado, Fig. 70. They are not in the aggregate of great extent, but they are important not only through their relief but also through their association with those great lava sheets that although greatly eroded still cover large portions of the Colorado Plateaus. It is important to see at the outset that the degree of dissection of the different portions of the province is not everywhere the same. The San Francisco Plateau is so little dissected that within its margins there is little hint of the great erosion forms to be found elsewhere in the region, and the flat plateau quality comes out strikingly in every general view; on the other hand the northeastern, north-central, and central portions of the province are dissected by a maze of ramifying canyons. The great breadth of the canyons in this portion of the province might be made the basis for a separate division of the pla- teau country— the Canyon Lands, as Powell termed the large dis- trict included in the valleys of the upper Colorado and the lower Green and Grand.!. So thoroughly dissected is this portion that the walls of the labyrinthine canyons are the dominating elements of the relief, and movement across the country is by tortuous and extremely toilsome routes, now paralleling some smali stream in the depths of a profound abyss, now crossing a plateau spur of mountainous proportions. The degree of topographic development which each district has attained will be the subject of further dis- cussion; here it is sufficient to note that although the strata in their horizontal development are decidedly uniform over the whole area, the topographic aspects of the various sections are as decidedly variable. A special feature of the plateau country is the manner in which ero- sion takes place. It is directed against the almost vertical edges of the strata much more than upon the almost flat upper surfaces. Plateau erosion is by stripping through cliff recession. Wash is almost ineffec- tive on the flat gradients of the plateau summits; on the steep and alternating slopes and cliffs of canyons and the cliffed edges of out- cropping strata everywhere it is vigorous, as is suggested by the name “Colorado,” whose colored, turbid current carries the waste of its 1 J. W. Powell, Lands of the Arid Region of the United States, U. S. Geog. and Geol. Surv. of the Rocky Mountain Region, 1879, p. ros. COLORADO PLATEAUS 259 great tributary system of » cliffs and slopes with their streams of land waste.! With this special erosion feature in mind it will be easy to understand how extensive the canyon systems of the plateau and the associated terraces are, and yet how. within the borders of each plateau there is in many cases so little dissection; each plateau is attacked prac- tically on its margins only | and so preserves almost until extinction a marked summit flatness. Conversely we should not understand be- cause such large expanses of flat plateau exist, and each is so little dissected within its borders, that erosion of the general surface is not active, for in the province as a whole, and between the flat-topped plateaus, erosion is taking place at a most rapid rate. The Colorado Plateaus consist of a large number of individual members sepa- rated from each other in various ways but in most cases by strongly defined lines, Plate II. The most striking line of division is the Colorado Canyon itself, which separates a northern The Looking southeast from Cortez. (Photograph by H. Brigham, Jr.) ‘ j 7 Formerly these beds extended farther toward the foreground; in time they will be pushed still 1 For a discussion of erosion by cliti recession see J. W. Powell’s various reports on the Colorado Plateaus. farther back. Sandstone caps the cliff and underlies the surface in the foreground. Between them are layers of shale. border of the mesa is a ragged cliff formed on Cretaceous beds. Fig. 71. — Panorama of Mesa Verde, southwestern Colorado; an example of mesa and scarp topography in the Colorado Plateaus. 260 FOREST PHYSIOGRAPHY from a southern series. Crossing the canyon almost at right angles are a number of northward-trending faults and associated fault scarps which block out an east-west series of plateaus. A third type of pla- teau boundary is exhibited in central Utah where the worn edges of outcropping, northward-dipping plateau strata constitute a line of remarkable cliffs of great physiographic interest and unusual scenic. beauty. Besides these well-marked divisions are others of lesser defi- nition. The northeastern and southeastern sections, although of great extent, have never been studied in the same detail as the rest of the plateau province, and their mutual boundaries are therefore less defi- nitely established. They are roughly separated from the other divisions by the Green-Colorado Valley on the northwest and the Little Colorado on the southwest, and from each other by the valley of the San Juan. The various lines of division block out four large and important districts or sub-provinces, a fact that needs emphasis, for the elementary student usually regards the great plateaus north of the canyon and in the Kaibab district as the only important parts of the province. The separate districts may be called for convenience: I High Plateaus of Utah: Awapa, Aquarius, Paria, Kaiparowits, Markagunt, Paunsdgunt, Tushar, Wasatch, Sevier, Fish Lake, etc. II Grand Canyon District: Shiwits, Uinkaret, Kanab, Kaibab, San Francisco, Coconino, etc. III Southern District: Zuni, Natanes, Taylor, etc. IV Grand River District: White River, Roan or Book, Uncom- pahgre, Dolores, etc. HicH PLATEAUS OF UTAH In any general view of the plateau country, as shown in Plate II, two distinctly different scarp systems may be seen. The north-south system as just described is caused by great faults with downthrow on the west. Almost equal to the north-south system of escarpments in magnitude is the east-west system that crosses the plateau country several hundred miles north of the Grand Canyon, a system due to plateau stripping, each cliff marking the outcrop of a resistant stratum. The individual scarps of the latter system break the northward conti- nuity of the Colorado Plateaus in a most decided manner and block out the country into a series of north-south plateaus, among which are the Colob, Markaégunt, and others, the whole series known as the High Plateaus of Utah.’ They are all characterized by great ruggedness of 1 C. E. Dutton, The Tertiary History of the Grand Canyon District and Geology of the High Plateaus of Utah, Mon. U. S. Geol. Surv., vol. 2, 1882. COLORADO PLATEAUS 261 outline, pronounced declivity, and a rude parallelism, although they are most irregular in detail. Named in order from south to north the prin- cipal ones are the Shinarump Cliffs, Vermilion Cliffs, White Cliffs, and Pink Cliffs, while many lesser cliffs block out plateaus of smaller extent. The cliffs and intervening plateaus constitute a group of great terraces from 30 to 40 miles in extent north and south and too miles east and west. Among the great cliffs perhaps the most remarkable in form are the White Cliffs, while among the most remarkable in color are the Vermilion Cliffs. The latter are from tooo to 2000 feet high, more than too miles long, and consist of evenly stratified layers of sandstone and shale with gypsiferous partings. In color they are brick red, which at twilight takes a strong vermilion hue; in form they are very ornate and architectural, with many vertical ledges rising tier above tier with intervening talus slopes through which the fretted edges of the cliffs pro- ject. Though the profile is complex it never loses its typical character and is always extremely picturesque because of the numberless alcoves and alternating promontories where streams cut into the edges of the plateaus which these cliffs terminate. The escarpments of the northern plateau country are all of a different type from those to the south. They consist of the outcropping edges of resistant northward-dipping strata that act as cliff makers and are being slowly stripped off the plateau surface by wind and water erosion. The former greater extent of the strata is inferred from the remnants left out upon the surface of the plateau south of the main cliffs in the form of isolated buttes and mesas. The map abounds in illustrations, of which perhaps the most conspicuous are the mesas that occur south of the Virgin River and west of Canyon Spring. In general the drainage of the region in which the east-west line of great cliffs occurs is from the north southward, and it is therefore in this direction that the main canyons run in contrast to the east-west course of the Grand Canyon farther south. There are three principal streams in this district: the Virgin flows west to the Great Basin and finally to the Colorado; Kanab Creek flows southward through a deep narrow gorge to enter the Colorado midway of the Kanab Plateau; and Paria River enters the Colorado at the head of the Marble Canyon. The beds of the plateau streams retain pools of water in the depressions pro- vided they are flooded with material that is not too coarse. These pools are called “water pockets,” “lakes,” “pools,” or “tanks.” They are scattered and few in number though always important features of the 1 C. E. Dutton, Tertiary History of the Grand Canyon District, Mon. U. S. Geol. Surv. vol. 2, 1882, pp. 17, 52, 53- 262 FOREST PHYSIOGRAPHY whole region. They are much more numerous in the higher plateaus of Utah than in the lower country. They are an important source of supply for travelers, settlers, and stockmen, and are the resort of bands of wild horses that roam the uninhabited desert tracts. The High Plateaus consist of three principal members, a western, a central, and an eastern, Fig. 72. The western member is composed of the Fig. 72. — Principal relief features of the High Plateaus of Utah. For Colb, lower left-hand corner, read Colob. Scale, 30 miles to the inch. (Dutton, U. S. Geol. Surv.) Pavant, Tushar, and Markagunt plateaus, named in order from north to south; the central member is composed of the Sevier and Paunsagunt plateaus, named in the same order; and the eastern member consists of the Wasatch, Fish Lake, Awapa, and Aquarius plateaus. COLORADO PLATEAUS 263 We shall describe only a few of the great plateaus of this northern region, selecting those which from our standpoint appear most impor- tant as types in the series. AQUARIUS PLATEAU The Aquarius, the grandest of all the High Plateaus, is about 35 miles long, of variable width, and 11,600 feet high. Its summit is clad with dense spruce forests sprinkled with grassy parks and exceptionally beautiful lakes.'. These are not small pools but broad sheets of water from 1 to 2 miles long. ‘Their existence is due to differential erosion by local glaciers that originated in the higher portions of the plateau; 8500 to gooo feet was the lower elevation of the glaciers of the region and it is at this level that the terminal moraines are usually found. The high elevation of the Aquarius Plateau, 10,500 to 11,600 feet, favored the exceptional development of glacial forms in the Pleistocene, just as to-day it favors a greater rainfall and better forest cover than occur on the neighboring lower plateaus. The upper surface of the Aquarius is developed upon a rooo- to 2000- foot layer of basalt, which gives rise to marginal cliffs of exceptional height and steepness, as on the northwestern flank. Again, on the eastern border, a great wall from 5500 to 6000 feet high overlooks the lower coun- try and owes its boldness largely to the hard lava cap at the summit.’ AWAPA PLATEAU The Awapa Plateau is 35 miles long and about 18 miles broad; its ele- vation is about gooo feet. The western border of the plateau is a wall from 1800 to 3000 feet high, but the other boundaries are less distinct. The slopes of the surface everywhere converge toward a central de- pression, Rabbit Valley. Although its altitude is that at which moisture and vegetation usually occur, only sage-brush and grasses grow and not a spring or a stream is found upon its entire surface. It is an endless succession of hills and valleys and shallow canyons (400 to 500 feet). It consists entirely of a great variety of volcanic material which has been poured out upon a sedimentary base or platform. Some of the grandest and most massive trachytic beds of the plateau region are found here. The irregular distribution and varied dissection of the many kinds of lavas that occur upon it are in large part the cause of the irregular relief of the surface.’ 1 C. E. Dutton, Geology of the High Plateaus of Utah, U. S. Geog. and Geol. Surv. of the Rocky Mountain Region, 1880, p. 5. 2 Idem, pp. 292-203. 3 Idem, pp. 272-276. 264 FOREST PHYSIOGRAPHY PARIA PLATEAU The Paria Plateau terminates on the south in a semicircular line of cliffs which are really a great southward prolongation of the Vermilion Cliffs and include the same strata (Triassic). It lies almost in line with the great Kaibab series of plateaus, yet topographically it is a part of the High Plateaus of Utah and like them is a great structural terrace. It is scored by a labyrinth of sharp narrow canyons which cut deeply into the platform-like surface. The course of the main drainage fea- ture, the Paria River, is independent of the dip of the strata; the courses of the smaller streams are all dependent upon: the structural dips.! Only the channel of the Paria, however, carries water. The rest are all dry channels which appear to have been formed during the moister glacial period and to have become functionless with the advent of the drier postglacial climate. Growing aridity has extinguished the smaller streams and increased the area drained by the living streams.’ KAIPAROWITS PLATEAU Between the Henry Mountains and the Paria Plateau, Plate II, is a broad area of plateau and canyon country known as the Kaiparowits Plateau and the Escalante Canyon. The canyons of both the Escalante and its numerous tributaries are a network of deep narrow chasms hemmed in by great unscalable cliffs. The depression carved by these streams is bordered on the north and west by a line of cliffs which ter- minate the Aquarius and Kaiparowits plateaus respectively. The Aqua- rius Plateau is forest clothed because high and relatively well watered; the Kaiparowits Plateau has only a scattered tree growth of very limited development; the depression below them on the southeast is waterless and treeless—a desert country, swept bare of soil. The cliffs which border the Kaiparowits Plateau on the northeast are 60 miles long and almost 2000 feet high. Their summit constitutes a divide from which almost no streams descend to the barren country below them on the east and only a few traverse the gentler western slope. MARKAGUNT PLATEAU The Markagunt Plateau, Plate II, is a broad plateau expanse south of the Tushar Plateau. It is limited on the west by the Hurricane fault scarp; the eastern base lies at the foot of the great Sevier fault on whose 1 C. E. Dutton, Tertiary History of the Grand Canyon District, Mon. U. S. Geol. Surv., vol. 2, 1882, p. 201. 2 Idem, p. 202. COLORADO PLATEAUS 205 eastern side the Pauns4gunt Plateau is found. The greater part of the area is covered with eruptive rock (trachyte) resting on sedimen- taries (Tertiary). The southern margin is a line of cliffs not made by faulting but by erosion of hard cliff makers in the northward-dip- ping series that constitutes the High Plateaus of Utah. The surface of the Markagunt consists of rolling hills and ridges and grassy slopes, the greater part covered with scattered groves of pine. Few canyons are sunk below its general level.' PAUNSAGUNT PLATEAU The Paunsdgunt Plateau is the southernmost extension of the High Plateaus of Utah and fronts south, with a marginal line of cliffs like its neighbors both east and west. These are the Pink Clifis, often resembling in a most striking way well-known architectural forms such as ruined colonnades, buttresses, and panels. Its strata are sensibly flat and are wholly of sedimentary origin, not volcanic flows capping sedimentaries as is the case with so many of the plateaus of this region.” TUSHAR PLATEAU The Tushar Plateau is highly inclined and is a transition type be- tween the flat plateaus general to the Colorado Plateaus province and the fault-block mountain type of the Great Basin. It is transitional in position as well as topography and structure, and, strictly speaking, lies within the geographic limits of the Great Basin. Its eastern front is steep and mountainous, for it has been developed largely across the edges of the strata; the western slope is down the dip of the strata and though considerably dissected is far less bold. Its summit is crowned by a cluster of peaks, true erosion remnants, which reach above timber line.’ WASATCH MOUNTAINS Although the Wasatch Mountains are not a part of the High Plateaus they are in such close relation to them as to demand a word of expla- nation at this point. They stand as a great wall upon the northwestern margin of the plateau country overlooking the Great Basin, and con- sist of a number of abrupt ranges crowned with sharp peaks that attain altitudes of 10,000 to 12,000 feet. Their boldness gives rise to a mod- erately heavy rainfall, and the mountain slopes bear forests of spruce, pine, and fir, while the broken and drier foothills support a growth of 1 C. E. Dutton, Geology of the High Plateaus of Utah, U.S. Geol. Surv., 1880, pp. 195 et seq. 2 Idem, pp. 251 et al. 3 Idem, p. 173. 266 FOREST PHYSIOGRAPHY Ltn, Pleasant Grove} | LEGEND GLACIAL DRIFT OF LATER EPOCH AGLACIAL DRIFT OF y /AEARLIER EPOCH PORTION OF SALT LAKE QUADRANGLE, U'TAIL Showing distribution of glacial formations : scale of Miles Oe ee 4 6 8 10 12 Contour interval 1000 feet Fig. 73. — Former glacier systems of the Wasatch Mountains. (After Atwood, U.S. Geol. Surv.) pinon pine and cedar. In the valleys there are many natural meadows but no forest growth, merely groves of aspen about the springs and lines of willow, box elder, and cottonwood on the borders of the streams.! ; The Wasatch Mountains extend about too miles south of the meeting point with the Uintas, or about to Mount Nebo, 75 miles south of 1 J. W. Powell, Lands of the Arid Region of the United States, U. S. Geog. and Geol. Surv., 18709, p. 96. COLORADO PLATEAUS 207 Great Salt Lake. Beyond this point the western margin of the plateau country is the western edge of the High Plateaus. These in turn give way to the broad platform of Carboniferous rock that constitutes the surface of the Grand Canyon District, whose western margin is the Grand Wash Cliffs. The eastern slope of the Wasatch Mountains falls off gradually as a 15 to 20 mile belt of broad ridges and mountain valleys whose waters reach Great Salt Lake through gorges that cut across the main western range of the mountains. The gentler eastern slopes are generally well clothed with vegetation. On the west the mountains present a bold abrupt escarpment which rises suddenly out of the broad flat plains of the Utah basin. The degree of abrupt- ness may be appreciated from the fact that the mountains attain elevations of 10,000 feet within 1 or 2 miles of the western base.! The main crest of the Wasatch Mountains is near the eastern border of the range and the western valleys are therefore much longer than the eastern valleys, — generally from two to three times as long. The loftier peaks — 11,000 to 12,000 feet — that are developed upon crys- talline or highly metamorphic rock have rugged, pinnacle-like forms, those developed upon horizontal sedimentary beds have pyramidal out- lines with alternating cliffs and talus slopes. In both cases the sharpness of form is due to glaciation. Summits that do not reach above gooo feet, and hence were never glaci- ated, are rounded and softened and bear a heavy cover of land waste. The elevation necessary for the development of Pleistocene glacier in the Wasatch Moun- tains was 8000 to gooo feet. Over 50 glaciers were formed exceeding a mile in length. Of these, 46 were west of the crest, and but 4 east of it. Of the ro exceeding 5 miles in length, 9 lay on the western slope, r on the _ ie eastern. ‘The greater number and size of the ee peat western glaciers were determined by the larger — Canyon, Wasatch Mountains. catchment areas and by the heavier snowfall, “t¥008: U: §: Geol. Surv.) the west slope being the windward or exposed slope. The western val- leys were therefore more completely cleared of loose material, the ex- posed surfaces more rounded, the main canyons deepened by a greater amount, and more massive moraines developed than in the eastern valleys. The typical relations of the moraines to each other and to Moraines of later, Moraines of earliers epoch : epoch 1S. F. Emmons, U. S. Geol. Expl. of the goth Parallel (King Surveys), vol. 2, 1877, p. 340. 268 FOREST PHYSIOGRAPHY the drainage features in a single valley are shown in Fig. 74. The ex- tent of the glacial systems and their relation to the topography and drainage are shown in Fig. 73.! GRAND CANyon DISTRICT The most celebrated and best-known portion of the Colorado Pla- teaus is the Grand Canyon district. The great north-south crustal fractures of this part of the plateau region are lines of faulting which block out the separate members of the district. Named in order from west to east the plateaus are the Shiwits, Uinkaret, Kanab, and Kaibab. Their elevations increase in the same order: the Shiwits is WEST Shivwits _ Uinkaret Kanab Marble Piatt? m ———., So = Tt N G.W. H. FT UO SEW KG Fig. 75. — East-west (top) and north-south (bottom) sections of the Colorado Plateaus, Grand Canyon district, showing both structure and topography. G.W., Grand Wash Cliffs; H., Hurricane Ledge; T., Toroweap fault scarp; W.K., West Kaibab fault; E.K., East Kaibab monocline; E., Echo Cliffs and monocline. (Davis, Bull. Mus. Comp. Zodl., modified from Dutton.) 5000 feet and the Kaibab, the highest of all, is 8000 to gooo feet above the sea. These several plateaus lie on the northern margin of the Grand and Marble canyons and constitute in the aggregate a great plat- form of nearly horizontal strata (Carboniferous) which is bounded on the east and north by the cliffs of the High Plateaus developed upon younger strata (Mesozoic) and on the west and south by its own termi- nal escarpment that descends to the older rock (Silurian and Archean) of the Great Basin and Arizona Highlands.? The elevations of all the plateaus except the Kaibab are not great enough to cause a precipitation adequate for a forest growth. In con- trast to the hot, dreary, and barren plateaus about it the Kaibab pla- teau is moist, and bears meadows and parks and forests of spruce and pine. The descent from one flat-topped plateau to another is over a 1 W. W. Atwood, Glaciation of the Uinta and Wasatch Mountains, Prof. Paper U. S. Geol. Surv. No. 61, 1909, pp. 73-03. 2 C. E. Dutton, Tertiary History of the Grand Canyon District, Mon. U.S. Geol. Surv., vol. 2, 1882, p. 19. COLORADO PLATEAUS 269 high, ragged, westward-facing cliff whose front is constantly battered back by wind and rain. The cliffs are great fault scarps which are strikingly linear and continuous. They are of such prominence as to cause a local rainfall which concentrates the erosive work of -the running water; detrital material, due to cliff recession principally, is washed forward from the foot of each cliff in the form of long, slop- ing, waste fans. The cliffs retreat without being obliterated, in part because of the horizontally bedded rocks, in part because of their recent origin and the small rainfall of the region. SHIWITS PLATEAU The Grand Wash Cliffs are the westernmost of the series in the Grand Canyon district and form the western border of the Shiwits Plateau. They are from tooo to 2000 feet high and overlook the Grand Wash and Huaipi valleys and the rugged sierras and stern deserts of the Great Basin. They form a continuous and bold line of cliffs from the upper Virgin Valley southward across the Colorado River almost 150 miles, to Music Mountain, where they are replaced by the Cottonwood and Yampai Cliffs. There is no river along this western wall of the plateau, ‘‘only occasional deluges of mud, whenever the storms from the southwest are flung against the lofty battlements and break in torrents of winter rain.””! The Grand Wash Cliffs mark the western end of the Colorado Canyon, 5000 feet deep; toward the west the Colorado suddenly changes its character and continues seaward without extraordinary features. Eastward of the Grand Wash Cliffs but still on the western border of the Shiwits Plateau is a second line of cliffs similar to the Grand Wash Cliffs in form but not in origin. It represents merely the eroded upper layers of the geologic series exposed by faulting and pushed back by ordinary erosion to their present position. The surface of the Shiwits Plateau is diversified by a number of vol- canic masses and a few large mesas and buttes, outliers of higher strata capped and so preserved by basalt. The most conspicuous height is Mount Dellenbaugh, 6750 feet high, a central mass of basalt in a large basaltic field.” UINKARET PLATEAU The Uinkaret Plateau, next east of the Shiwits, is separated from the latter by Hurricane Ledge, a fault scarp tooo feet high which is con- 1 C. E. Dutton, Tertiary History of the Grand Canyon District, Mon. U.S. Geol. Surv., Vol. 2, 1882, p. 12: 2 Tdem. 270 FOREST PHYSIOGRAPHY tinued south of the Canyon by the Aubrey Cliffs, of similar height and origin.! The Uinkaret Plateau is the narrowest of the four and its southern portion is the most strongly diversified by basaltic eruptions. In- deed this portion (20 miles from north to south) is so broken as to be known as the Uinkaret Mountains in contrast to the Uinkaret Plateau farther north In sharp contrast to the rectilinear outlines and vivid colors of the greater part of the plateau region are the irregular profiles and gloomy aspect of the basaltic plateau of Mount Trumbull, a great mesa consisting of horizontal sedimentary strata capped by 500-600 feet of basalt. Its summit (2000-3000 feet relative altitude) reaches to such a height that the climate is cool and moist and the plateau sustains a forest growth of yellow pine. About the mesa or mountain and covering the whole southern end of the Uinkaret Plateau are great lava flows — jagged masses of black basalt, desolate except for an occa- sional grove of cedar and pinon.® From the summit of Mount Trumbull 120 to 130 cinder cones are visible and upon the whole Uinkaret are 160 to 170 vents in all. None of them in the main field is of great size; the highest are only 700 or 800 feet in altitude, with a diameter of a mile. KANAB PLATEAU The Kanab Plateau is separated from the. Uinkaret by an incon- spicuous boundary. For about 20 miles north of the canyon it is clearly marked by the line of cliffs along the Toroweap fault which grad- ually dies out northward, and beyond this point no prominent topo- graphic feature forms a dividing line.* The Uinkaret and Kanab plateaus are separated from each other by the twenty-mile long Toroweap Valley, the locus of the Toroweap fault that causes the eastern wall of the valley to be several hundred feet higher than the western. The greatest displacement along the line of the fault is only 700 feet. It extends but 18 or 20 miles north of the canyon, hence is a much weaker boundary than the neighboring faults. 1 The heights of the various fault scarps should not be taken as an indication of the amount of faulting. While the Grand Wash Cliffs are from 2000 to 3000 feet high, the fault which gave rise to them exhibits 6000 feet maximum vertical displacement. Likewise Hurricane Ledge is from 500 to 2000 feet high, but it is associated with a maximum displacement of about 12,000 feet 10 miles north of the Virgin River Valley. (Dutton, idem, p. 20.) 2 See the Mount Trumbull quadrangle, U. S. Geol. Surv. 3 C. E. Dutton, Tertiary History of the Grand Canyon District, Mon. U.S. Geol. Surv., vol. 2, 1882, pp. 81-84, 103. 4 Idem, p. 13. COLORADO PLATEAUS 271 KAIBAB PLATEAU The Kaibab and Kanab plateaus are separated by the West Kaibab fault. The eastern border of the Kaibab Plateau is not formed mainly by faulting, hence it is an exception to the general rule. Here occurs not a fault but a great monoclinal flexure that is partially and locally faulted on a small scale and with a total descent of about 4000 feet. About mid-length it becomes a double monocline and finally a double fault before reaching the Colorado. The whole displacement dies out south of the Colorado. The Kaibab Plateau is thus a great uplifted block between two parallel lines of displacement, the eastern a com- pound flexure, the western a fault. The western border fault continues southward for but a short distance before splitting into three secondary faults with associated scarps and these extend almost to the brink of the Grand Canyon.! The northern end of the Kaibab is a great cusp which terminates a little south of the town of Paria. Its southern border is the mile-deep Grand Canyon of the Colorado. Structurally and topographically the Kaibab is continued across the canyon into the southern district, but the strong break of the canyon has caused the adoption of a different name for the southern section, which is called the Coconino Plateau. It would be reducing to common terms that which is in the highest degree uncommon if no mention were here made of the extraordinary scenic features displayed in this greatest wonderland of North America. Even superlative terms are feeble, and ordinary language is wholly in- adequate for the expression of forms and colors that have been the theme of every enthusiastic scientist and every poet who has beheld the great cliffs, profound canyons, and vast expanses of the plateau region. Dutton’s rich vocabulary enabled him to write a description which is probably the most satisfactory expression of the ideas and feelings which these great scenic features inspire. We therefore restrict ourselves to two or three quotations from his classic memoirs, Tertiary History of the Grand Canyon District and Geology of the High Plateaus of Utah. They are not the most enthusiastic passages that may be found; they are rather the most restrained and careful and may therefore be accepted not only as scientific but also as imaginative and interesting. The first description presents the region of terraces and canyons east and south of the High Plateaus of Utah as seen from one of the latter at an altitude of more than 11,000 feet. 1 C. E. Dutton, Tertiary History of the Grand Canyon District, Mon. U. S. Geol. Surv., vol. 2, 1882, p. 13. 272 FOREST PHYSIOGRAPHY “. . the eye ranges over a vast expanse of nearly level terraces, bounded by cliffs of strange aspect, which are truly marvelous, whether we consider their magnitude, their seem- ingly interminable length, their great number, or their singular sculpture. They wind about in all directions, here throwing out a great promontory, there receding in a deep bay, but con- tinuing on and on until they sink below the horizon, or swing behind some loftier mass, or fade out in the distant haze. Each cliff marks the boundary of a geographical terrace sloping gently backward from its crest line to the foot of the next terrace behind it, and each [in northward succession] marks a higher and higher horizon in the geological scale as we ap- proach its face. Very wonderful at times is the sculpture of these majestic walls. Panels, pilasters, niches, alcoves, and buttresses, needing not the slightest assistance from the im- agination to point the resemblance; grotesque forms, neatly carved out of solid rock, which pique the imagination to find analogies; endless repetitions of meaningless shapes fretting the entablatures are presented to us on every side, and fill us with wonder as we pass.””! ‘“‘Isolated masses cut off from the main formation, and often at considerable distances from it, lie with a majestic repose upon the broad expanse of the terrace. These sometimes become very striking in their forms. They remind us of great forts with bastions and scarps nearly a thousand feet high. The smaller masses become regular truncated cones with bare slopes. Some of them take the form of great domes where the eagles may build their nests in perfect safety. But noblest of all are the white summits of the great temples of the Virgin gleaming through the haze. Here Nature has changed her mood from levity to religious solemnity, and revealed her fervor in forms and structures more beautiful than anything in human art.” 2 “But of all the characters of this unparalleled scenery that which appeals most strongly to the eye is the color. The gentle tints of an eastern landscape, the rich blue of distant moun- tains, the green of vernal and summer vegetation, the subdued colors of hillside and meadows all are wanting here, and in their place we behold belts of fierce staring red, yellow, and toned white, which are intensified rather than alleviated by alternating belts of dark iron gray.’’3 “As the sun nears the horizon the desert scenery becomes exquisitely beautiful. The deep rich hues of the Permian, the intense red of the Vermilion Cliffs, the lustrous white of the dis- tant Jurassic headlands are greatly heightened in tone and seem self-luminous. But more than all, the flood of purple and blue which is in the very air, bathing not only the naked rock faces but even the obscurely tinted fronts of the Kaibab and the pale brown of the desert surface, clothes the landscape with its greatest charm. It is seen in its climax only in the dying hours of daylight.”’ 4 SAN FRANCISCO PLATEAU The term is here employed to include the whole region between the Grand Wash Cliffs, the Arizona Highlands, the Grand Canyon Dis- trict, and the Little Colorado River. It includes the Coconino Pla- teau on the north, and on the south the broad platform of much greater extent upon which the San Francisco Mountains stand.’ Its general 1 Geology of the High Plateaus of Utah, p. 8. 2 Description of the scenery of the White Cliffs, developed upon the white Jurassic sand- stone. From Tertiary History of the Grand Canyon District, p. 37. 3 From description of the same region noted in the first quotation from Geology of the High Plateaus of Utah, p. 8. 4 Impressions of a sunset on the Kanab Plateau in Tertiary History of the Grand Canyon District, pp. 124-125. 5 It is somewhat confusing to use the term Colorado Plateau for this portion of the district as in Dutton’s standard description. The term Colorado had already been established for the entire province. Both the Land Office maps and the maps of the Topographic Branch of the U. S. Geological Survey follow Dutton’s usage however. In using the term San Francisco Plateau I am following the excellent suggestion of Robinson (Am. Jour. Sci., vol. 24, 1907). COLORADO PLATEAUS altitude is from 7000 to 7500 feet, which is distinctly higher than the plateaus im- mediately north of the canyon, except the Kaibab. The strata of the district are nearly horizontal though with a broad re- gional slope toward the southwest, and, with few exceptions, the surface is not deeply or extensively scored by canyons. Unlike the Kaibab it has no strongly marked divisional boundaries, since the great fault scarps which form such promi- nent features of the one are either absent or diminish in height or disappear in the other. Its surface is diversified in some localities by (1) low, lava-capped, and usu- ally forest-clad mesas, as Black Mesa and Mogollon Mesa, and (2) volcanic cones and peaks such as San Francisco Mountains (13,000) and the associated lava flows about their bases... Some of the volcanoes are very recent, as in the San Francisco Moun- tain region, an example 16 miles north of San Francisco Mountain being probably not more than too or 200 years old.? SOUTHERN PLATEAU DISTRICT ZUNI PLATEAU The Zuni Plateau (locally called the Zuni Mountains) is an illustration of a type of structure seen in a number of places in the plateau country, for example in the San Rafael Swell on the eastern margin of the High Plateaus, and is of special physi- ographic interest. The otherwise flat pla- teau strata have been somewhat locally 1 C. E. Dutton, Tertiary History of the Grand Canyon District, Mon. U. S. Geol. Surv., vol. 1, 1882, pp. 14-15. 2 D. W. Johnson, A Recent Volcano in the San Fran- cisco Mountain Region, Arizona, Bull. Geog. Soc. Phil., vol. 5, 1907, pp. 2-6. FROM LITTLE COLORADO TO CAMP APACHE Mogollon Mountains Basaltice Aubrey, Cherty Limestone Red Wall- Limestone Lava and Stratified Sandstone Arenaceous Shales 273 Vig..76. — Topographic and structural section of the Mogollon Mountains and Mesa. Typical border topography southern edge of Colorado Plateaus, see Plate II. (Modified from Marvine, Wheeler Surveys.) 274 FOREST PHYSIOGRAPHY uplifted, raising the surface well above the general level and exposing it to pronounced erosion. The cross section, Fig. 77, exhibits the main structural features. It will be readily seen that the margins of the uplifted tract have been structurally bent and in places faulted, but that the summit of the uplift is almost flat. The erosion of the superior part of the broad uplift has removed portions of the capping layers and exposed the geologic section ranging from rather young (Cretaceous) to very old rocks (Archean). Each stratum ends in a cliff facing the central axis of the uplift, and each cliff may be traced in the form of a great oval entirely around the tract. Each stratum also dips outward from the cliff summit and extends below the next cliff of the younger formation overlying it. The heart of the uplift is composed of Archean gneisses and mica schists, and these constitute the main summit platform or plateau over a considerable portion. Since the summit has this constitution the oldest (stratigraphically the lowest but topographically the highest) sedimentary formation, the Car- boniferous, ends at the shoulder or margin of the main platform. Many portions of the summit have strongly marked topo- graphic features and are very rugged and diversified, being deeply scored with canyons. The principal drainage lines cross portions of the uplift regardless of the structure and show that they were determined before the uplift, a feature of common occurrence throughout the plateau country, while it is equally common to find all the lesser stream-ways in close sympathy with the broad structural surfaces of the main plateaus.! MOKI-NAVAJO COUNTRY?” That part of Arizona bounded by the San Juan, the Colorado, Little Colorado, and Puerco rivers reveals in slightly modified form the typical characteristics of the plateau province. The river valleys are canyons or wide, open washes, the rocks are chiefly sedimentaries of Mesozoic age, the climate is arid, the vegetation sparse, and there is little prospect that the area can be made profitable to civilized man. The only permanent streams are the San Juan, the Colorado, and the headwaters of 1 C. E: Dutton, Mount Taylor and the Zuni Plateau, 6th Ann. Rept. U. S. Geol. Surv., 1884-1885, pp. 162-163. 2 In the absence of published data on this section of the plateau country Prof. H. E. Gregory has kindly prepared this brief description from his field notes. Fault Section from northeast to southwest across the Zui Plateau. The strata are Archean, Carboniferous, Jura-Trias, and Cretaceous, in upward succession. (Dutton, U. S. Geol. Surv.) Fig. 77. COLORADO PLATEAUS 275 a few rivers which come from the highlands; for example, Chinlee, Moencopie, Navajo, and Little Colorado, Plate II. During the rainy season (July and August) the run-off from the high- lands is so rapid and uncontrolled and the flat valley washes are so thoroughly drowned that travel is exceedingly difficult and often danger- ous; and the utilization of the extensive alluvial bottoms for agriculture on a large scale is rendered impossible. None of the important streams is reduced to grade throughout its course. All of them are carrying prodigious quantities of waste to be added to the burden of the Colo- rado. In the absence of running streams the Navajos and Hopis depend for themselves and their stock upon springs and ephemeral lakes, and the supply at best is meager and commonly unpalatable. It is re- markable how little water is used by man and beast. The broad structural features of the region are easily comprehended. The highlands along the Arizona-New Mexico line include Carrisos Mountains at the north, the Lukachukai, Tunichai, and Choiskai moun- tains, and the hogback ridges between Fort Defiance and Gallup. The Carrisos (elevation 9420 feet) rise 5000 feet above the San Juan at their base, and constitute one of the most prominent landmarks of the pla- teau country. The mountains are laccolithic in origin and are caused by irregular intrusions of quartz-monzonite which deformed Triassic, Jurassic, and Cretaceous sandstone into a group of domes. Subse- quent erosion has removed the cover here and there, exposing the igne- ous core, and deeply trenched the heavy sedimentary beds which form the flanks of the uplift. Lukachukai, Tunichai, and Choiskai moun- tains are composed of sedimentary strata lying approximately hori- zontal; their elevation is due to a fold of great amplitude, the western limb of which forms the eastern side of the Chinlee Valley. Caps of lava rest upon the Tunichais as well as upon a number of detached buttes at the south, while the covering beds of the Choiskais are Ter- tiary sediments. The ridges south and east of Fort Defiance are formed by the resistant members of Mesozoic strata which here dip to the east at a high angle. The area between the lower Chinlee and Colorado rivers is part of a great dome of Triassic and Jurassic sediments intersected by the San Juan River. At Marsh Pass the strata plunge beneath the Creta- ceous of the great Zilh-le-jini (Navajo for Black Mountain) mesa, and the hogback rim of the dome is prominently exposed from Marsh Pass, Arizona, to Elk Ridge, Utah. The elevation at Skeleton Mesa and the region about the head of Pahute Canyon is attained by a series of mono- clinal folds with steep eastern limbs. As one travels from the Chinlee 276 FOREST PHYSIOGRAPHY westward he ascends a series of giant steps until an elevation of nearly 8000 feet is reached. Standing on the plateau at this elevation his view is unobstructed except for Navajo Mountain (10,400 feet), a lacco- lith with its sedimentary cover still intact. It stands as an island covered with vegetation in the midst of a sea of barren red rock. From the top of Navajo Mountain a more comprehensive view of the plateau province may be obtained than from any other point. From Lee’s Ferry the Leupp Echo Cliffs and their southern continuation mark the line to which the Jurassic and Upper Triassic have been stripped by the tributaries of the Colorado and the Little Colorado. In the center of the Navajo Reservation and including most of the Hopi Reservation is the Zilh-le-jini mesa, attaining an elevation of 6000 to 8000 feet, and bordered on all sides by steep escarpments. In struc- ture this highland is a shallow syncline of Cretaceous strata in which are included valuable beds of coal. On the narrow finger-like mesas forming the southern border of Zilh-le-jini are located the six villages of the Hopis which constitute the ancient province of Tusayan. Be- tween Tusayan and the Little Colorado is an extensive area of lava- capped mesas and volcanic necks and dikes, the remnants of early Tertiary lava flows. The vegetation is characteristic of the plateau province. Above 7000 feet on the highlands along the Arizona-New Mexico boundary line there grows an open stand of yellow pine which is used for the manu- facture of rough lumber for various government projects. Pine in lim- ited amounts grows on the northern rim of Zilh-le-jini mesa and covers the upper slopes of Navajo Mountain. Between 6000 and 7000 feet, pifon and cedar are commonly found, but over wide areas it averages not more than two or three trees per acre. Sage-brush and greasewood grow in limited amounts at elevations below 6000 feet, and grass, which is fairly abundant at higher elevations, is also found in limited amounts below 6000 feet, and in a few favorable localities forms a continuous sod. The extent of bare rock and sand floor is very great, and probably not more than 10% of the Navajo-Moki region is actually covered with vegetation. In fact it is possible to walk from Gallup, New Mexico, to Tanner’s Crossing on the Little Colorado, or from the Carrisos Moun- tains to Lee’s Ferry, without stepping on a twig or a spear of grass. GRAND RIVER DISTRICT West of -the Rocky Mountains of central Colorado, Plate II, and on meridian 107° 30’ is the northeastern edge of the Colorado Plateaus. This portion is called the Grand River district after the main drainage COLORADO PLATEAUS 277 line. It consists of the White River Plateau, Grand Mesa, the Book or Roan Plateau, the Uncompahgre Plateau, and the Dolores Plateau, named in order from north to south. The strata upon which these plateaus are developed as a rule dip toward the west and away from the Rocky Mountains. In the southwestern part of Colorado, along the San Juan Valley, six well-marked groups of hard rock (sandstone) occur and a corresponding number of soft shales alternate with them and form lines of weakness for the attack of the weather. Each group of strata slopes and dips gently westward until a break occurs and a precipitous descent is made across the edges of the rock layers to the next lower group of strata,! a succession that is typical of the district, although the number of alternations of strata and to some extent the topographic expression varies from section to section. Like the other districts the Grand River district exhibits a number of mountain groups of igneous origin scattered upon its plateau surfaces. These include the San Miguel, La Plata, Carriso, Abajo, and other groups, which are as a rule of laccolithic origin. Erosion has removed the once overlying sedimentary beds and exposed the intruded rock. WHITE RIVER PLATEAU The White River Plateau is the northernmost of the series in the north- eastern part of the plateau province. The general level of its surface is interrupted in many places by higher summits (500-1000 feet above the general level) and ranges of mountains on the one hand and by deep valleys and canyons on the other. While plateau surfaces are still conspicuous features of the region, the upland has been so largely cut away by erosion, so largely modified by uplift or faulting of the once flat-lying strata, so complicated by intruded masses of igneous rocks, that the plateau feature can scarcely be said to be the dominating one, and by some the plateau would be classified as a part of the Rocky Mountains. The best-preserved sections of the plateau are those whose flat upper surfaces have been unbroken and but slightly tilted. The White River Plateau has been formed largely by lava flows, the borders of which have been cut into deep gorges and ravines in which some of the headwaters of the White and Yampa rivers take their rise.2 On the west especially this plateau falls off in abrupt and high cliffs to the long slopes of the bordering spurs; the eastern border is marked by high detached masses which give it a mountainous aspect 1 C. E. Dutton, Mount Taylor and the Zuni Plateau, 6th Ann. Rept. U.S. Geol. Surv., 1884-1885, p. 242. 2 F. V. Hayden, U. S. Geog. and Geol. Surv. of the Terr., 1876, p. 7. 278 FOREST PHYSIOGRAPHY when viewed from the east. North of the White River Plateau the country is mountainous and so broken that no distinct, well-defined topographic system has been determined.' On the south the plateau forms the divide between the White and the Grand rivers and loses its distinctive mesa-like quality, being cut by profound canyons.” SPECIAL BORDER FEATURES In portions of northwestern Colorado the plateau character is largely replaced by hogback or monoclinal ridges, strike valleys, and other fea- tures related to moderate complexities of structure developed in a series of alternating hard and soft strata.’ One of the most remarkable physiographic features of northwestern Colorado is Grand Hogback, which extends southward from the Dan- forth Hills to a point beyond Grand River. Indeed under the name of Colorado Ridge it continues far beyond Grand River to the point where its identity is lost in the more rugged West Elk Range. The Grand Hogback is bordered on both sides by long continuous valleys; the ridge itself is at some places a single-, at others a double-crested hog- back ridge formed by the outcrop of massive sandstone ridges inclined eastward at steep angles. Its steep inclination has afforded opportunity for the development of an exceedingly rugged topography whose principal elements are precipitous marginal slopes. The most striking features of the Grand Hogback and the Colorado Ridge are their topographic continuity and structural uniformity. The stream valleys in the Hog- back Ridge contain a small amount of red fir, while the upper slopes and crest, 2000 feet above the marginal valley bottoms, are covered with a dense growth of oak brush, chokecherry, and juniper.* ROAN OR BOOK PLATEAU One of the most striking topographic features of the Grand River dis- trict is the southern margin of the Roan or Book Plateau, the Roan or Book Cliffs.° In places the cliffs rise almost vertically from the edge of the Grand River Valley to their full height; elsewhere the ascent is by a succession of broken steps. Their mean height is about 8600 feet, or 1S. B. Ladd, U. S. Geog. and Geol. Surv. of Colorado and Adj. Terr. (Hayden Surveys), 1874, PP. 437-438. 2 Idem, p. 439. 3 H. S. Gale, Gold Fields of Northwestern Colorado and Northeastern Utah, Bull. U. S. Geol. Surv. No. 415, 1910, p. 24. 4 Idem, p. 31. 5 Roan, from their prevailing color; Book, from the resemblance of their characteristic form to a bound book. COLORADO PLATEAUS 279 3500 feet above the Grand River. They are the southern escarpment of the plateau of the same name which slopes rather gently to the north and northeast and is drained by tributaries of the White River. Nearly half the original surface of the plateau is still intact, so that a rather flat aspect is more common than in the more dissected White River Plateau on the east or the Uncompahgre Plateau on the south. The western is the most dissected portion of the plateau, and here the divide is in places only 30 or 40 feet wide, bordered on the cliff side (south) by a precipice and on the plateau side by a strong slope. The crest of the cliffs has very little water on account of the low elevation (8600 feet); it is covered in the main with grass and sage. Almost the only arboreal vegetation is quaking aspen, which occurs but sparingly; there are but a few occurrences of spruce and pine.! ~ UNCOMPAHGRE PLATEAU The Uncompahgre Plateau lies between the Uncompahgre Mountains and the Grand River. It is 90 miles long and from 15 to 25 miles wide; its elevation is from 8600 to 10,200 feet (northwest). It is in the form of a broad arch of sedimentary beds about tooo feet thick over a cen- tral core of granite, the latter being well exposed in the steep and beauti- ful canyon of the Unaweep, where it forms two-thirds of the height of the walls. The tributaries of the Unaweep have all cut down to the granite, and in the season of floods due to the melting of the snows their waters drop from 300 to 2000 feet to the bed of the main canyon below. On the west the plateau is bordered by steep cliffs which descend to the valleys or canyons of the Grand and other streams. One of the most profound canyons of the whole mountainous western half of Colorado is that of the Gunnison on the northeastern border of the Uncompahgre Plateau. It is 56 miles long and 3000 feet deep in the deepest part. The plateau in which it is incised consists of granite- ' gneiss capped by rooo to 1200 feet of sedimentary strata. The canyon is cut through the overlying sedimentary rock and deep into the gneiss, the contact being marked by a sloping bench below which are rough, ragged, and nearly vertical walls which extend to the river margin; above the contact are steeply sloping talus and vertical cliff in alternating series. The tributaries of the Gunnison have incised their courses but little into the gneiss and therefore have very steep descents where they join the master stream.” 1 F. V. Hayden, U. S. Geog. and Geol. Surv. of the Terr., 1875, p. 346, and 1876, p. 60. 2 Henry Gannett, U. S. Geol. and Geog. Surv. of Col. and Adj. Terr. (Hayden Surveys), 1874, Pp. 425. 280 FOREST PHYSIOGRAPHY The influence of elevation upon rainfall and thus upon the character and amount of the vegetation is admirably illustrated in the Un- compahgre Plateau. In the interior of the plateau and down to an elevation of 7000 feet or more are streams and springs in some number, good pasturage, a sprinkling of aspen groves, and game in some quantity; below 7000 feet aspen gives way to pinon, grass to sage, cacti, and bare rock, the streams become muddy torrents or dry up altogether, and in place of the rolling plateau surface are deep narrow canyons and steep precipices.! The influence of elevation upon the character of the vegetation is also well shown on the plateau between the Gunnison ‘and the North Fork of the Gunnison. This plateau has a smooth unbroken summit sloping in a direction slightly west of north. It ranges in elevation from gooo to 5400 feet, and owing in part to its marked slope, which drains the surface water rapidly away, and in greater part to its low elevation, the plateau is without timber in contrast to the bordering mountains of slightly greater elevation. Its vegetation is pinon pine, cactus, sage- brush, and scrub oak in contrast to the timbered plateau on the east.? DOLORES PLATEAU The Dolores Plateau lies between the Dolores and San Miguel rivers about halfway between the San Juan and Grand River canyons. The highest portion is known as Lone Mesa and is at 10,000 feet eleva- tion, with an area of about 40 square miles. About it are a number of high flat-topped buttes which like it are erosion remnants with steep bordering scarps. The superior elevation of Lone Mesa gives it a greater rainfall than falls on adjacent tracts and meadows, and forests of pine abound.® The canyon of the Dolores is very narrow and precipitous, with almost no alluvial bottoms except in its shallowest portion at the great bend, Fig. 78. Here are a rich growth of grass, some cottonwood groves and bushes, and vines. On the borders of the canyon below this point is a rather heavy growth of pinon pine and cedar, and on the various head- water branches are forest and meadow, aspen groves, and rich grassy parks in contrast to the desert canyon below.’ One of the most remarkable cases of stream direction not in accord with the present slope of the plateau surface (p. 284) is that of the 1F. V. Hayden, U. S. Geog. and Geol. Surv. of the Terr., 1875, pp. 340, 341, 340. 2 Henry Gannett, U. S. Geol. and Geog. Surv. of Col. and Adj. Terr. (Hayden Surveys), 1874, Dp. 426. 3 Idem, p. 266. 4 Idem, pp. 266-277. COLORADO PLATEAUS 281 Dolores River, a tributary of the Grand. The Dolores rises in the San Juan Mountains, runs south and west for more than 30 miles, then flows northwest against the inclination of the surface for about 60 miles to the Grand River in western Colorado, in a gradually deepening canyon, Fig. 78. At the turning point it is in a canyon only too feet deep; above and below this point the canyon deepens to 2000 feet. Evidently the river gained its course sometime before the present surface con- ditions were established. The surface is structural in origin, being developed on a great sandstone layer (Dakota). Either the surface was peneplaned and the river now pursues an antecedent course with re- LONE CONE oe MTS. Sa pancos- ZN je Fig. 78. — Canyon of the Dolores. The canyon is 2000 feet deep at a, 2100 at b, and roo at c. (Hayden Surveys.) spect to later uplifts that deformed the peneplain, or its course was developed in response to a structural slope that has since been reversed. While the main stream has thus maintained its earlier course the smaller streams of the region show marked responses to the present attitude of the surface. The tributaries of the San Juan (south) rise almost on the brink of the Dolores canyon, for they have been extended in response to the structural surface now existing and have encroached on the weaker Dolores tributaries until the latter are all but extinguished south of the canyon.' The Yellow Jacket tributary of the San Juan has encroached so far upon the Dolores at the great bend that a tunnel several hundred yards long has been constructed through the in- tervening barrier and a large part of the water of the Dolores turned into the Yellow Jacket for the purpose of irrigating the valley flats at Cortez, Plate IT. PHYSIOGRAPHIC DEVELOPMENT; EROSION CYCLES An understanding of the physiography of the Colorado Plateaus re- quires at least some knowledge of its recent history as expressed in a number of topographic cycles through which the region has passed. With the long pericds of deposition in the plateau region when it stood 1F. V. Hayden, oth Ann. Rept. U. S. Geog. and Geol. Surv. of the Territories, 1875, pub. 1877, pp. 263-264, and Plate 42, p. 264. 282 FOREST PHYSIOGRAPHY at or below sea level we have little to do; nor is it necessary for an un- derstanding of existing topography to take account of the ancient and now buried surfaces of erosion that are so well exposed in the walls of the Grand Canyon. It is sufficient for our purpose to begin with rather late movements in the evenly bedded mass of strata accumulated by long-continued erosion of the Great Basin and adjacent mountains. In the Tertiary (latter part of the Eocene or possibly in early Miocene) the plateau region was uplifted and the uplift was accompanied by monoclinal folding. The result of these first deformations was an elevation of the western plateau country above the eastern, and the descent from plateau to plateau was at that time from west to east and not as at present from east to west. The folding probably gave rise to a number of closed basins in which lakes were formed and sediments laid down, although the drainage of the region was probably on the whole through basins draining to the sea. Since this first deformation of the plateau country (in Tertiary time) the region has suffered continuous erosion down to the present, although the character of the erosion, the geologic structure, and the topography, have been modified repeatedly, as outlined in the following paragraphs. In response to internal forces of the earth a period of faulting followed the first period of monoclinal folding and by it the plateau district was marked out and the eastern and western borders clearly defined through the lowering of the country on either side. The date of this period of faulting has not been closely ascertained, but it may be regarded with a certain degree of accuracy as having occurred at the close of the Miocene. FIRST EROSION CYCLE The two geologic events of monoclinal folding and of faulting lent to the relief of the plateau region a pronounced character. But the topographic expression of these structural features was gradually and Fig. 79. — Black Point Monocline, Colorado Plateaus, showing remnant of base-leveled surface capped by lava. 1. Kaibab; 2. Moencopie; 3. Shinarump; 4. Basalt. (Robinson). at last completely obliterated at the close of a cycle of erosion known as “The Period of the Great Denudation.”’ The surface of the entire southern plateau country, and probably also of the northern plateau country, was reduced to a peneplain. At the close of the great denuda- tion cycle there occurred widespread eruptions of basalt. The lava flows capped portions of the surface of the peneplain and thus protected them from the effects of later denudation. An instructive locality for the study of those features upon which the foregoing description rests is at Black Point in the Little Colorado Valley, where the Black Point monocline dips eastward as shown in the cross section, Fig. 79. The black basalt which caps the surface of the country and preserves a portion of the ancient peneplain rests upon an exceedingly smooth, almost flat surface, a degree of smoothness that could not have been developed across strata COLORADO PLATEAUS 283 of such variable hardness except at a base level of erosion. The Kaibab cherty limestone is very resistant, while the Moencopie soft sandstone and shales and the Shinarump marls are very soft and easily weathered; above the marls at the eastern end of the section is a compact sandstone of most resistant quality. A similar occurrence is at Anderson Mesa, eight miles southeast of Flagstaff, where a smooth surface was once developed across rocks of very differ- ent- hardness such as the Kaibab cherty limestone and the soft Moencopie shales. The peneplain is also well shown in the Mount Taylor region on the walls of both the Mount Taylor and the Prieta mesas where it bevels across beds of slight dip. Its character as in- ferred from the once buried remnants now reéxposed was that of a surface of rather faint relief.t Huntington and Goldthwaite? have describedit in the Toquerville district, Utah, and Davis has interpreted phenomena of the same significance in a number of other localities. The boundaries of the known portions ‘of the base-leveled surface in- clude about 25,000 square miles of country, and it seems at present as if these peneplain remnants in various portions of the plateau country were at one time united so as to form a surface of very slight relief, out of which the existing plateaus were blocked by faults. Recently the work of Gregory* in the Navajo country and the valley of the San Juan has shown the extension of the great peneplain surface far to the north. An excellent locality is between Tuba City and Oraibi, where a remarkably flat surface only partially dissected bevels regularly across strata of pronounced dip and structural variation. The entire extent of the peneplain will only be known when the now little explored portions of the province (which aggregate the greater part of it) are examined and the physiography interpreted. SECOND EROSION CYCLE The next important geologic event in the plateau region was the inauguration of a second period of faulting near the close of the Pliocene, a period of faulting in which the plateau region was again strongly blocked out and given those features that are most prominent at the present time. The second period of faulting increased the relief of the plateau region and resulted in the pronounced step-like relation of the different members of the plateau. While the folding of the first period of deformation operated in such a manner as to cause a descend- ing series of plateaus from west to east, the major faults of later dates reversed the order of descent, an order that has been maintained down to the present. The second period of faulting introduced the second or post-peneplain cycle of erosion, in which there was developed a widespread system of 1 —D. W. Johnson, Volcanic Necks of the Mount Taylor Region, New Mexico Bull. Geol. Soc. Am., vol. 18, 1907, pp. 307-308. See also idem, cross section, Fig. 2, p. 300. 2 Huntington and Goldthwait, The Hurricane Fault in the Toquerville District, Utah, Bull. Mus. Comp. Zodl., vol. 42, Geol. Series, vol. 6, 1903. 3 Personal communication. 284 FOREST PHYSIOGRAPHY shallow but mature valleys, one of the most persistent features of the better-known portions of the province. The drainage characteristics of this partial cycle have been noted by several writers, among whom Robinson was the first to demonstrate their persistence and their meaning in terms of erosion cycles.'!. The same feature has been de- scribed by Noble,? who says: “The drainage system of the plateau surface (Coconino) consists of a series, of mature open- floored valleys with gently sloping sides, which contain no living streams. . . . In tracing one of these mature valleys towardits head . . . [itis] common. . . to find it truncated as a hanging valley by the wall of the Grand Canyon. . . . The same system of mature valleys covers [the] surface [of the Kaibab plateau] which slopes southwesterly to the rim of the canyon.” The surface drainage of the Kaibab now runs through this system of mature valleys into the Grand Canyon; the surface drainage of the Coconino runs through a similar valley system away from the Grand Canyon, and both plateaus have only temporary streams. These mature drainage systems offer the most striking contrast on the one hand to the youthful topography of the deeply trenched canyon developed in the third or next erosion cycle, and on the other hand to the base- leveled surface now preserved in fragments beneath the basalt caps of various mesas. It should not be thought that the second cycle of erosion brought about any great vertical reduction of the surface. On the contrary the depth to which the streams cut in response to the new base level was only a few hundred feet as against the few thousand feet of the third and last, or canyon cycle of erosion. The chief result of denudation in the second cycle was the broad horizontal stripping back of the strata outcropping on the surfaces of the plateaus and the development of mature valley systems. The stripping of the gently inclined strata proceeded on structural planes, so that the present flat plateau surfaces are structural surfaces developed upon the upper surface of resistant strata. Generally the structure dips uniformly toward the southwest at the rate of about 200 feet to the mile, so that the surfaces of the Coconino and San Francisco plateaus south of the Grand Canyon slope to the southwest from the canyon rim at the rate of about 200 feet per mile. The Kaibab plateau north of the Colorado Canyon slopes southwesterly in the same degree to the rim of the canyon. It is for this reason that the southern plateau drains away from the canyon and that the northern plateau drains into it. The plateau surfaces are 1 H. H. Robinson, A New Erosion Cycle in the Grand Canyon District, Arizona, Jour. Geol., vol. 18, 1910, pp. 742-763. 2 L. F. Noble, Contributions to the Geology of the Grand Canyon, Arizona, The Geology of the Shinumo Area, Am. Jour. Sci., vol. 29, 1910, pp. 374-380. COLORADO PLATEAUS 285 everywhere accordant with the rock structure.' These facts must be thoroughly appreciated because, while the plateau as a whole has been base-leveled, the remnants of the ancient base-leveled surface total a very small area, and in general are preserved only beneath lava caps near the summits of mesas. THIRD EROSION CYCLE The third period of faulting and of regional uplift inaugurated a third cycle of erosion, and, because the most striking effects of erosion in this cycle are the great canyons of the region, it has been called the “Canyon Cycle of Erosion.” During this cycle a certain amount of plateau stripping has taken place and the cliff profiles have been re- freshed, although the chief result has been the development of the profound canyons of the great Colorado and its principal tributaries. During the canyon cycle of erosion there occurred a third period of volcanic activity char- acterized by eruptions of basalt from many small volcanic cones. The later eruptions took place, as a rule, before or during the period of glaciation, since they have been glacially modified, although a few cones and flows are of very recent geologic age. The later basaltic flows may be distinguished from the earlier flows capping the surface of the peneplain by their freshness, their undissected condition, and the absence of those displacements which affected the earlier volcanic flows.? With these historical events in mind we may now turn to some fea- tures of the existing topography of the Colorado Canyon that require Fig. 80. — Cross-profile of the Grand Canyon of the Colorado River. Scale, one inch = 15,000 feet. I, inner gorge; 2, 5, 7, sandstone; 4, 8, limestone; 3, 6, shale. (Gilbert and Brigham.) more detailed study. The main section of the canyon is known as the Grand Canyon. Above the Grand Canyon is the Marble Canyon (cut into Carboniferous limestone) a feature of great magnitude but dwarfed by the adjacent highly diversified Kaibab division of the Grand Canyon, more than a mile deep (6000 feet). The length of the Marble Canyon is 65 miles, that of the Grand Canyon about 125 miles. The two main divisions of the Grand Canyon are the Kaibab and the Kanab, which have certain topographic contrasts of importance. In 1 L. F. Noble, Contributions to the Geology of the Grand Canyon, Arizona, The Geology of the Shinumo Area, Am. Jour. Sci., vol. 29, 1910, pp. 374-380. 2 H.H. Robinson, Tertiary Peneplain of the Plateau District and Adjacent Country in Arizona. and New Mexico, Am. Jour. Sci., vol. 24, 1907, pp. 110-112. 286 FOREST PHYSIOGRAPHY the Kaibab division of the canyon the descent of the wall is unusually abrupt over the entire series of sedimentary rock. The only well- defined terrace is that developed upon the summit of the basal sand- stone of the Tonto group near the bottom of the canyon—~a terrace known as the Tonto platform. Below the platform is an inner gorge formed upon the basement schists (Algonkian). The platform aver- ages a mile wide on either side of the canyon and is well enough defined to make travel upon it possible. In the Kanab division to the west the lower terrace has disappeared and a different terrace has been developed upon the summit of the Red Wall limestone about a thousand feet below the level of the canyon rim. Through it as through the Tonto platform is cut a deep narrow inner gorge. This platform or terrace is known as the Esplanade; it averages two miles in width on either side of the canyon. Still further differences may be seen in the aspect of these two sections of the canyon. The topography of the Kaibab division exhibits much greater dissection than that of the Kanab. The former is diversified by great amphitheaters thronged with buttes and outliers and trenched by a multitude of tributary gorges. In the Kanab division a very simple topography is found. There is a fairly regular broad outer canyon in which is cut an inner gorge, and the fantastic scenery of the Kaibab division is here wholly absent. CLIMATIC FEATURES AND VEGETATION The distribution of temperature and rainfall according to relief or elevation is brought out strikingly in the plateau province. An inch Colorado Plateaus San Francisco Lower Colorado Arizona Basin Highlands he Prescott \ \ 1 / 7S ‘ a 3S 2.7 in. Sea Level Fig. 81. — Topographic profile in relation to rainfall distribution from southwest to northeast across the three physiographic provinces of Arizona. of rainfall is the normal increase for every 500 feet of altitude on the border of the plateaus; about half this increase in altitude is required for an increase of an inch of rainfall within the border of the plateaus. Yet most of the plateaus are dry, almost as dry as the plains of south- western Arizona at half the altitude above sea level. The dryness is explained chiefly by the effect of the abrupt southwestern and western border in draining the passing winds of water vapor. 1 L. F. Noble, loc. cit., p. 375. COLORADO PLATEAUS 287 Below an altitude of 7000 feet the rainfall of the Colorado Plateaus is probably not more than 8 inches a year. With increase of altitude above that figure there is increasing rainfall and the highest of the High Plateaus receive the equivalent of about 24 inches of rain, though a large part of it occurs in the winter months and is mostly in the form of snow.t. They are favored however with a summer maximum in July and August. Indeed Dutton estimates their summit rainfall at 30 inches.2, The High Plateaus are therefore not arid; they are the first prominent topographic barrier which the winds strike east of the Sierra Nevada. The natural vegetation of the Rio Grande Valley on the southeastern border of the plateau country is limited to scanty grass, cottonwoods, and willows on the river bottoms, and to cacti and other desert forms on the higher slopes. At still greater elevations the junipers begin to come in, at first as gnarled stunted growths, then in better form, and on the slopes of the volcanic mountains, as Mount Taylor and the San Francisco Mountains, and on the summits and to some extent on the slopes of the higher mesas such as Mogollon Mesa, Black Mesa, and the Zuni Plateau are thousands of square miles of magnificent forests of yellow pine and spruce.? This is the usual succession of vegetation in the plateau country. The altitudinal range of timber species is sometimes not clearly defined on account of diversity of climatic conditions and exposure. At the headwaters of the East Clearwater, a tributary of the Little Colorado, the shady northward facing canyon slopes and walls are timbered with red fir and white fir, while the drier southward facing slopes are covered entirely with yellow pine.t| The most valuable tree of the region is the yellow pine; the sugar pine is found only on the southern plateaus and is rarely of commercial size; the pifon pine and the pitch pine are common but not valuable growths and usually occur on talus slopes. Engelmann spruce occupies the highest elevations and is the only timber above 11,000 feet, attaining its best growth at 1o,ooo feet and disappearing at 11,500 feet. The great height at which it grows makes it difficult to secure; and to the altitude is added the difficulty of the 1 A. H. Thompson, in Powell’s Lands of the Arid Region, U.S. Geog. and Geol. Surv. of the Rocky Mountain Region, 1879, p. 151. 2 C. E. Dutton, Geology of the High Plateaus of Utah, U. S. Geog. and Geol. Surv., 1880, p. 41. 3 C. E. Dutton, Mount Taylor and the Zuni Plateau, 6th Ann. Rept. U.S. Geol. Surv., 1884- 1885, p. 125. 4 F. G. Plummer, Forest Conditions in the Black Mesa Forest Reserve, Arizona, Prof. Paper U.S. Geol. Surv. No. 23, 1904, p. 16. . 288 FOREST PHYSIOGRAPHY broken canyoned country between its plateau and mountain home and the valleys where the people live.! The growth of grass is in most places scant, and at lower elevations diminishes in quantity and disappears in the lower and more desert country except where springs occur. The grasses grow characteristically in bunches and are thus in part protected among themselves from wind and blown sand. They have large strong stems and are not easily broken down by the infrequent rains and snows. They cure on the stalk and are highly nutritious, furnishing winter pasturage of great value to stock- men.” The range of climate between the summits of the plateaus and the lower desert country of the canyon terraces and canyon bottoms is great indeed. The high precipitation of the Aquarius Plateau, for ex- ample, 25 to 30 inches, is chiefly in the form of snow which accumu- lates in the forest to a great depth. Settlers find it very difficult to live at an altitude over 7000 feet; their farms are usually found below that level where the climate is hot, arid, or semi-arid, and where irrigation is a necessity. From the cool, lake-besprinkled forests and meadows of the higher plateaus one looks down upon a country of extreme heat and dryness. The range of climate between the two situations is as great as one may find in most regions only by traveling through a considerable number of degrees of latitude, — a common range, however, in the great intermontane country between the Rockies and the Pacific Mountains. The range in climate between the Kaibab and the bottom of the canyon, for example, is as great as the climatic range between the moun- tains of Colorado and the Mohave desert. The winters on the Kaibab Plateau are extremely severe and from November until April the snow lies deep in the woods, often accumulating to a depth of ro feet. Even in midsummer the nights are chilly and the days are cool. The winters in the depth of the canyon are mild, freezing temperatures are rare, and snow rarely falls below the level of the Esplanade (4500 feet), while snow never falls on the Tonto platform (2100 feet).’ In contrast to the cool summer days of the plateau is the intense heat of the entire canyon below the Red Wall. The bare rocks become so hot as to burn the hand, and by nightfall a wind like a furnace blast 1 J. W. Powell, Lands of the Arid Region of the United States, U. S. Geog. and Geol. Surv., 1879, pp. 98-103. See these pages for a more extended description of the tree species of the plateau province, their habitats and relative commercial importance at the time of Powell’s surveys. 2 J. W. Powell, Lands of the Arid Region of the United States, U. S. Geog. and Geo!. Surv. of the Rocky Mountain Region, 1879, p. 110. 3 Nokle, loc. cit. COLORADO PLATEAUS 289 escapes through the canyon. But the heat is not enervating, for the relative humidity is very low and moisture is rapidly evaporated from the body. The bottom of the canyon is decidedly arid, for much rain- fall evaporates before it reaches the great depths. The Coconino Plateau on the south side of the canyon has a lesser altitude than the Kaibab Plateau, and while the latter is decidedly moist the former is semi- arid. Often no rain falls for a month at a time. In both cases there are distinct summer and winter maxima. Powell Plateau on the south- western border of the Kaibab has a high eastern portion, and a low western portion, and from end to end there is a transition in climatic conditions similar to the transition that is experienced in passing from the Kaibab to the Coconino Plateau. In winter it is a resort for game and wild horses driven out of the Kaibab by the snow. The higher eastern end has an abundant rainfall; the lower western end is semi-arid. One of the finest forests of the whole region is found on the high and better-watered Kaibab. The trees of the Kaibab forest are mostly yellow pine, but at the higher elevations spruce also is common. Pines are found on the sunny slopes of the ravines, spruce on the shady slopes, and both grow only scatteringly upon the valley bottoms. These and the aspens are the three principal genera, but about the borders of the plateau are patches and scattered individuals of cedar (Juniperus occidentalis), mountain mahogany (Cercocaspus ledifolius), and pifion.' The influence of elevation upon vegetation through temperature and rainfall is admirably shown in the Grand Canyon, where it has been observed by Noble, from whose excellent descriptions the following paragraphs are taken?: “The surface of the Kaibab Plateau is covered with a magnificent open forest of yellow pine; the trees grow large and far apart and the ground is free from undergrowth, giving its surface the aspect of a great park; Engelmann spruces grow on the north slopes of the washes, and cottonwoods, aspens, and scrub oaks in their bottoms; a minor flora of flowering plants, exceedingly rich in species, covers the floor of the forest. The flora of the plateau surface or the south rim of the canyon differs completely from that of the Kaibab; it is covered with a forest of gnarled and stunted trees of juniper and pinon, with here and there a buckbrush bush; the trees never form thickets, but grow wide apart; while the open stretches are covered with sagebrush and mormon tea, with occasional cactus, mescal, and plants of the century family. This difference between the floras of the north and south rim is due to the differences in pre- cipitation and temperature, which vary directly with the altitude. For this reason the floras of the plateaus-furnish an almost unfailing index of the elevation. This is beautifully shown on the southwestward-sloping surface of Powell Plateau, the whole eastern half of which lies at an elevation of from 7000 to 7500 feet and is covered with the open pine forest character- istic of the Kaibab. At about 7000 feet the character of the flora changes, and passes into the 1 C. E. Dutton, Tertiary History of the Grand Canyon District, Mon. U. S. Geol. Surv., vol. 2, 1882, p. 132. 2 L. F. Noble, Contributions to the Geology of the Grand Canyon, Arizona (The Geology of the Shinumo Area), Amer. Jour. Sci., vol. 29, 1910, pp. 374-380. 290 FOREST PHYSIOGRAPHY gnarled and stunted forests of juniper and pinon characteristic of the southern plateau across the canyon. “Within the canyon itself the variation in the flora is just as great, and is again an index of the elevation. “The flora of the Esplanade platform, a thousand feet below the south rim, consists of stunted bushes of juniper and pifion, with greasewood as the ground bush in place of the sagebrush of the Coconino Plateau. The cactus, mescal, and plants of the century family are present in greater abundance than on the plateau, but in less abundance and in more stunted develop- ment than in the bottom of the canyon. This is due to the fact that the Esplanade level is within reach of the winter snows and frosts. ‘ “The flora of the Tonto platform, three thousand feet below the south rim, and of all the interior of the canyon below the Red Wall, is the flora of a hot and arid desert in its most characteristic form. The dominant plants are the greasewood bush, the mormon tea, and the cactus. The mescal and the plants of the century family here attain their greatest develop- ment and size. The cacti are particularly rich in species. Every plant in the flora is either prickly or aromatic; leaf surfaces are reduced to a minimum; devices for storing water attain the greatest perfection; and the dominant color is a somber gray. The somber colors and the reduction of leaf surface are apt to deceive the observer, both in regard to‘the richness of the flora in species and the abundance of plant life, which is far greater than one would suspect. The only tree is the screw-mesquite, which grows in the beds of those washes that contain liv- ing or intermittent streams. “The vegetation in the bottoms of those canyons of the north side in the Shinumo Amphi- theater which contain living streams is beautiful beyond description, and in refreshing con- trast to the desert flora of the Tonto platform. Tall cottonwoods grow in the lower canyons; the walls are hung with maidenhair fern in the shady places; and willow thickets border the stream. Grass grows on the banks where there is soil. Higher up in the canyons, caks, maples, and other deciduous trees come in, and often beds of tall rushes. The most character- istic bush of these upper north-side canyons is the manzanita, which does not grow on the south side of the Grand Canyon.” In the northeastern or Grand River district there is a characteristic change in vegetation on ascending from the valley or canyon floors to the plateau summits. In the low valleys and on the dry ridges sage- brush occurs; in the moist valleys are found willow, buffalo berry, service berry, and along the larger streams cottonwood. The quaking aspen requires more water and is found only above 7500 feet, on the plateau summits or in the vicinity of springs or on cool and moist slopes. All the low bluffs and ridges support some pifon and juniper of low height growth, yellow pine occurs infrequently, and what is locally called white pine (Abies engelmanni) is found only in some of the gulches and ravines leading down from the summit of the Book Cliffs.! MOUNTAINS OF THE PLATEAU PROVINCE HENRY MOUNTAINS The mountains of igneous origin that occur locally throughout the plateau country have been referred to in a number of preceding para- graphs in this chapter. Some of the most important features of the 1 F, V. Hayden, U.S. Geog. and Geol. Surv. of the Terr., 1876, pp. 68-69. COLORADO PLATEAUS 291 Fig. 82. — Relief map of the Henry Mountains, a group of eroded laccolithic domes in the plateau country, southeastern Utah. (Gilbert, U. S. Geol. Surv.) larger groups deserve a word of detailed explanation. Our first con- sideration will be the Henry Mountains, which, with the San Francisco and Mount Taylor groups, have been studied in more detail than others of their kind. The Henry Mountains are in southern Utah and are on the right 292 FOREST PHYSIOGRAPHY bank of the Colorado River between the Dirty Devil and Escalante tributaries. They are a group of five individual mountains separated by low passes. Although they are in an arid region their height (7000 to 11,000 feet) is such as to cause them to have a rainfall sufficient to support forests. Mount Ellen is the highest of the five mountains and has a continuous crest line 2 miles long, with radiating spurs and bor- dering foothills. The Henry Mountains were formed by the intrusion of great masses of molten rock into surface beds in the form of a laccolith. The overarching strata that were lifted up as a conse- quence of the intrusion have been so extensively removed by erosion that there is now revealed the heart of the laccolith, and the displaced beds are exposed on the flanks and borders of the mountains for the most part. There is considerable diversity in the degree of erosion; some of the mountains are still largely covered on one or more sides by overlapping sedimentary rocks. The Henry Mountains in southern Utah stand upon a desert plain having a mean altitude of 5500 feet. A large part of the rain that falls in the region is distributed over the mountains, a fact that is re- flected in the distribution of the springs and the vegetation. On the surrounding plain the vegetation is extremely meager, grasses and shrubs and in favored localities the dwarf cedar (Juniperus occiden- talis). The highest peak of the group is Mount Ellen, 11,250 feet, which bears cedar about its base, pifon a little higher up, then yellow pine, spruce, fir, and aspen. The summits are naked; the upper un- timbered slopes are covered with a luxuriant growth of grasses and herbs. Of the forest growths noted the pines grow ina scattering man- ner, but the cedars and the firs are in dense groves.!. Among the other four mountains of the five comprising the group, Mount Pennell (11,150 feet) reaches above timber line and has a forest growth similar to that on Mount Ellen except that the timber reaches almost to the summit; on Mount Hillers (10,500) the timber reaches to the principal summit, but is less dense than on the other higher mountains; Mount Ellsworth is so low (8000) that it bears none of the forest trees that grow on the others, its vegetation is cedar and pifion right to the summit, and the grasses are less rank and grow in scattered bunches; Mount Holmes (7775) has a similar growth except that a few spruce trees grow high up on the northern flank. Among the extinct volcanoes and associated lava flows Mount Taylor, San Francisco Mountain, Sierra Mogollon or the Mogollon Mesa, the Panguitch Lake Buttes of Utah and the Marcou Buttes of New Mexico are Important members. These have all been formed by the extrusion 1G. K. Gilbert, Report of the Geology of the Henry Mountains, U. S. Geog. and Geol. Survey of the Rocky Mountain Region, 1877, p. 118. COLORADO PLATEAUS 293 of lava in one or another form, either as volcanic mountains or as lava flows. Like the laccolithic mountains they have no range forms or structures. Their features are all grouped about centers of structural disturbance, and radiating slopes and spurs and drainage lines are the rule. They have the utmost differences of position with respect to timber lines. Many do not reach above the dry timber line, some do not reach the cold timber line, a few extend above the cold timber line and have unforested summits. Those having a forest cover at all are en- circled by a band of forest vegetation from the 7000 or 8000 foot level to the 11,000 foot level, which represents the upper limit of tree growth. In all of them there is a notable canting of the timber lines on passing from the warmer and drier southern slopes to the colder and moister northern slopes. The difference of level amounts in some instances to a thousand feet and is always several hundred feet, being lower on the northern slopes. This condition has been especially well described by Merriam.! More recently Lowell has noted that the degree of canting increases with increasing altitude and appears to be a function of radiation and insolation as controlled by area. With increasing altitude there is de- creasing area. The two sides of a cone will therefore show increasingly greater differences in the limiting elevations of the tree zones and a more marked canting of the vegetation belts and timber lines.? | zl SNH E ET. SAN FRANCISCO PEAKS LAT. B52 20EN. CRATER HILL Agassiz Humphrey LAT. 35°14'N. WING MTI. Alpine Zone Timber Line Zone White Spruce » Douglas Fir -P.in esstatatatenes te SSSRANS ||| Pinon TUTTI ie f Dotted Line Shows the Crater F = Flagstaf 1 Horizontal Division of Scale =2 Miles Q=The Observatory 1 Vertical 7? | 7» |=2000 Ft-=5.28 Horizontal Fig. 83.— Timber zones on San Francisco Peaks, Ariz., showing increase in the degree of canting with increasing elevation. (Lowell.) 1 C. H. Merriam, Results of a Biological Survey of the San Francisco Mountain Region in Arizona, North Am. Fauna, No. 3, 1800. 2 Percival Lowell, The Plateau of the San Francisco Peaks in its Effects on Tree Life, Bull. Am. Geog. Soc., vol. 43, ro11, p. 380. 204 FOREST PHYSIOGRAPHY SAN FRANCISCO MOUNTAINS AND MOUNT TAYLOR SAN FRANCISCO MOUNTAINS The San Francisco Mountains are the center of a volcanic field in northern Arizona that includes seven large peaks and several hundred small peaks. Cones and lava flows together cover about 2200 square miles of country. San Francisco Mountain, the highest peak in the group, rises to 12,700 feet above the sea and has 5000 feet of relative altitude. Other prominent peaks are Kendrick Mountain, 10,500 feet, and Bill Williams, Sitgreaves, Elden, and O’Leary mountains, which do not exceed 9500 feet. The smaller volcanic cones are scattered about irregularly and have great variety of topographic detail. The plateau about the base of these mountains and throughout the volcanic field maintains a greater altitude than elsewhere, a fact due to (1) greater initial height of the locality before volcanic activity began and (2) the protective influence of the hard lavas which have preserved the surface from that denudation which later brought the surrounding plateau to a lower level. San Francisco Mountain is a composite cone built up by the lavas and breccias of five dis- tinct periods of eruption. The chief vents of these periods are near each other, so that the total effect has been to give the mountain a symmetrical outline. The principal lavas are dacite, rhyolite, and andesite. Twelve principal watercourses drain radially outward from the higher parts of the area. They are irregular in the lava field, but be- come more direct beyond the borders of the field, where deep canyons generally occur. After a cloud-burst or heavy shower water runs for a few hours in the canyons or washes within the limits of the storm area and then the channels become dry again. But one stream, Oak Creek, runs throughout the year; its more even plow is due to several large forest springs at its headwaters. In an area of 1000 square miles about San Francisco Mountain only about 25 springs and water pockets or ‘“tanks’’ occur, and at least two-thirds of these are situated near the mountain, so that the surrounding country is extremely dry and difficult to traverse. The “tanks” and lakes which usually occur at the heads of the washes have a most variable supply as regards both quantity and quality. The lakes are due to the damming of the watercourses by lava flows. Some of them have become dry by the cutting down of their outlets and the accumulation of sediments; their floors are now grass-covered and picturesque glades. The soils of the San Francisco volcanic field vary from “adobe” soils of the floors of temporary lakes and ponds to the pervious cinder COLORADO PLATEAUS 295 and scoriaceous soils of the slopes of volcanic cones and ridges. Be- tween these two extremes are gravelly loams of several varieties moder- ately pervious to water and best adapted of all the soils to the growth of forests. The coarse cinder soils lose water so rapidly that they are extremely sterile, while the adobe soils crack open when dry and swell when wet and are not adapted to forest requirements.! The San Francisco Mountains are encircled by barren plateau country, but they are themselves covered with a beautiful forest of juniper, pine, Aa a Fig. 84. —Mesa forest of western yellow pine in the Mogollon Mountains near Iron Creek, Gila National Forest, New Mexico. (Photograph by DeForest.) and spruce. It surrounds the base of the mountains and _ stretches westward to the borders of the escarpment of the Colorado Plateaus and down it nearly to its foot. The forest accompanies the escarp- ment southwestward to the boundary of Arizona and extends in a north- west-southeast direction for over 200 miles. Its greatest breadth is opposite the San Francisco Mountains, where it is nearly 50 miles wide. Elsewhere its breadth is from 12 to 25 miles, thus giving it an area of from 12,000,000 to 23,000,000 acres. It is from all points of view the finest forest in the Southwest and is composed of an almost pure growth of yellow pine except in the higher altitudes of the San Francisco Moun- 1 Leiberg, Rixon, Dodwell, and Plummer, Forest Conditions in the San Francisco Moun- tains Forest Reserve, Arizona, Prof. Paper U. S. Geol. Surv. No. 22, 1904, p. I5. 296 FOREST PHYSIOGRAPHY tains. It is an open forest with little undergrowth and the trees are of good size for lumber purposes.! The yellow-pine zone extends from 7000 to 8200 feet. Above it and between elevations of 8500 and 9800 feet and occasionally to 10,000 feet is the transition forest type, composed principally of red fir, white fir, aspen, and a few Engelmann spruce. The subalpine forest type extends from 9800 to 12,400 feet and is found only on Kendrick Moun- tain and San Francisco Peaks. It is composed of aspen, Engelmann spruce, etc. Above the subalpine forest is a treeless belt~which is rep- resented only on and near the summit of San Francisco Peaks, Fig. 83.” Below the yellow-pine zone and between 5700 and 6200 feet is a woodland belt where both climate and soil are very dry. Juniper and pifon are the chief species. They stand as a transition type of vege- tation between the desert sagebrush and the true forest of the next higher belt. MOUNT TAYLOR Northeast of the Zuni Plateau is Mount Taylor one of the most promi- nent volcanic elevations of the southeastern section of the province. The central mass and associated lava flows constitute the summit of a mesa whose base consists of sedimentary rock. The mesa is 47 miles long and has an extreme breadth of 23 miles, with an average altitude of about 8200 feet. Mount Taylor itself is 11,390 feet high. It is a lava cone principally. and was a central pipe or vent from which the sur- rounding flows were derived. It is now very much eroded, though it still maintains a roughly amphitheatral form. It is heavily timbered and deeply covered with soil and talus. In the Mount Taylor district and east of the main volcano are many lesser volcanic elevations ‘now greatly denuded so that only the central core remains. They stand from 800 to 1500 feet above the general level of the plain as steep-sided sharp-crested buttes of great interest as the roots of once active volcanoes now all but swept away.’ Among the other areas of higher relief in the Colorado Plateaus is Mogollon Mesa, a timbered plateau 7000 feet above the sea and extend- ing southward from the San Francisco Mountains (Plate II). Limestones and shales form the basement rock upon which the basaltic lavas that 1 Henry Gannett, roth Ann. Rept. U. S. Geol. Surv., 1897-08, p. 47. 2 Leiberg, Rixon, Dodwell, and Plummer, Forest Conditions in the San Francisco Mountains Forest Reserve, Arizona, Prof. Paper U. S. Geol. Surv. No. 22, 1904, pp. 18-10. 3 C. E. Dutton, Mount Taylor and the Zuni Plateau, 6th Ann. Rept. U. S. Geol. Surv., 1884-85, especially Panorama in the Valley of the Puerco, Plate 21, p. 171; also D. W. Johnson, Volcanic Necks of the Mount Taylor Region, New Mexico, Bull. Geol. Soc. Am., vol. 18, 1907, PP. 303-324. COLORADO PLATEAUS 207 constitute the mass of the mountains have been distributed. These strata dip northward and their edges are exposed on the southern border of the district where the lava cap has been most extensively frayed by dissection, Fig.83. Here the main margin of the elevation consists of long lava-capped promontories of the plateau fronted by a series of detached remnants in the form of mesas and buttes. The mountains have been extensively dissected and present a bold and varied relief in contrast to the more regular margin of the mesa above which the mountains rise as from a pedestal. Thrifty forests of yellow pine cover both the mesa and the higher slopes and owe their existence to the favorable tem- perature and rainfall induced by the elevated position of the moun- tains on the windward border of the plateau country.' 1 Wheeler Surveys, Geology, vol. 3, 1875, pp. 217-587. CHAPTER XVIII ROCKY MOUNTAINS. I NORTHERN ROCKIES BOUNDARIES AND SUBDIVISIONS BETWEEN the sensibly flat and only slightly dissected basalt plains of the Columbia River and Snake River regions and the almost equally flat Great Plains of Montana is a broad belt of wild, rugged, mountain- ous country composed of many northward- and northwestward-trend- ing mountain ranges. Though several transcontinental railway lines cross the region and occasional valleys, as the Bitterroot, are rather thickly settled, the forested mountain slopes and the higher ranges have few trails and fewer roads, and are eos unknown except to the hunter and the explorer.' The chief members of the northern part of the Rocky Mountain System? in the United States are the Lewis and Livingston ranges, which form the eastern and front members of the system in northern Montana; the Galton, Flathead, Purcell, Cabinet, and other ranges east of the continental divide; the central Bitterroot Mountains on the boundary between Montana and Idaho; and the Clearwater, Salmon River, Priest River, and Cceur d’Alene mountains on the west, Fig. 86. The close proximity of these ranges to each other, in general their similar align- ment and their common participation in certain important events in the physiographic history of the region, are conditions which form an adequate basis for the common association of the separate units under the group name of the northern Rockies of the United States; but there are wide differences in local detail, in the quality of the moun- tain slopes, the length and disposition of transverse valleys, and the degree of dissection which the individual ranges have suffered. It there- fore becomes necessary to distinguish the essential elements of form and the limits of each unit in the system. 1 F. C. Calkins, A Geological Reconnaissance in Northern Idaho and Northwestern Mon- tana, Bull. U. S. Geol. Surv. No. 384, 1900, p. 12. 2 The United States Geographic Board makes the term Rocky Mountain System embrace the whole of the mountainous region between the Rio Grande and the 4oth parallel, specifi- cally the ranges of western Texas, New Mexico, Colorado, Wyoming, Idaho, and Montana. In this book the Trans-Pecos Highlands are set off as a separate province. (See p. 387.) 298 So I5u0 ‘ ‘ INvy 3avo Pend Oreh (ie) Great Falls Ey Y, Col faxe€ C| Snake Lewi ‘al «a eh oWallawalla AK Zl 118° 7 Fig. 85. — Mountain systems and ranges and intermontane trenches, northern Rockies. (Ransome, U. S. Geol. Surv.) 112 a Sun [River Scale of Miles wt Ni or, co) cS \ So ta Lewistown Plateau 13° hd ~ Ag roy 5 orn [= 4 Creal, iy KG oO Co Vr, Bitteroo / Lever id P, ay Lire posit Salm >| i>, y Mowstune— Stinky FyLSLake Tam Woe S0/SE mrg\NOF : Fig. 86. — Location map of a part of the Northern Rockies. 99 300 FOREST PHYSIOGRAPHY In naming and describing the different mountain ranges of the north- ern Rockies that cross the international boundary line it has been pro- posed that a double and purely topographic principle be employed, the principle of the continuity of crest lines and the positions of, the major erosion valleys.' Erosion troughs or intermontane trenches are here the natural lines of demarcation between the different ranges, for they have a highly exceptional and remarkable development on a great scale. INTERMONTANE TRENCHES In respect of size the major valleys of the region are often quite out of proportion to the streams that now drain them; the longest depression in the whole system is occupied by relatively small streams, the head- waters of the Kootenai, Columbia, Frazer, and others. They have steep though seldom precipitous walls, and rather flat, lake-dotted floors. They are called intermontane or valley trenches. It is certain that they had their direction determined in many cases primarily by fault lines or zones, so that some of them, for example Kootenai Valley, mark the boundary between different rock formations. They have probably been brought to their present form chiefly by the long-con- tinued erosion of valley glaciers powerful enough to override the divides between the heads of the larger streams and to degrade them to a common level.? In this connection it is in point to indicate that valley glaciers of great size supplied by many headwater tributaries are much more powerful agents of erosion in effecting topographic discordances such as appear in the hanging valleys and the mountain slopes adjacent to steep-sided U trenches, and in opening up valleys such as these trenches are to-day, than are continental ice sheets. A continental ice sheet is spread over the entire surface, and though it may erode to some extent differentially there is no such concentration of ice action: as is the case in a region of heavy alpine glaciation. The valleys of 1 R. A. Daly, The Nomenclature of the North American Cordillera between the 47th and 53d Parallels of Latitude, Geog. Jour., vol. 27, 1906, pp. 586-606. This excellent paper out- lines the difficulty of understanding the mountain nomenclature of the Rockies, and on pages 589 and 590 summarizes the various names applied by prominent authorities to the Rocky Mountains. In all at least 26 different names have been employed. It is suggested that greater differentiation be aimed at in designating the various members of the Rocky Mountain system. It is also noted that the authorities variously designate the terminal points of the different mountain ranges (pp. 592-503). The nomenclature is still further confused owing to the fact that some atlases and map sheets give different titles to the same range. 2 F. C. Calkins, A Geological Reconnaissance in Northern Idaho and Northwestern Mon- tana, Bull. U. S. Geol. Surv. No. 384, 1909, p. 32- ROCKY MOUNTAINS. I 301 Lake Chelan, the Okanogan, and the Cceur d’Alene are illustrations. In all three instances there are extensive catchment basins which fed prodigious glaciers. Among the eleven longitudinal valleys or intermontane trenches which serve as convenient lateral boundaries for the members of the Pacific Cordillera are four of first rank; three of these occur in the northern Rockies, and lie west of the front ranges, the Purcell range and the Selkirks respectively. Each is designated a “trench,” and as so used the SEEKIRK-SYSTEM PURCELL TRENCH —W———_—__COEUR D'ALENE MTS. TR =<—Rathdrum) Rtaltie => eens eau Pend Oreille_Lake in SpahaneVal ieyS = Coeur_D? Alene Lake 4 Horizontai Scale Late fa md Ge 0 1 2 3 4 5 miles (Vertical Scale but slightly exaggerated) Fig. 87. — East-west section, showing flat floor and steep bordering slopes of a typical intermontane trench. (Data from Cceur d’Alene quadrangle, U. S. Geol. Surv.) term means ‘‘a long, narrow, intermontane depression occupied by two or more streams (whether expanded into lakes or not) alternately drain- ing the depression in opposite directions.” ! The easternmost and much the longest of the series of trenches is the wide valley occupied at the international boundary by the Kootenai River.? It extends from the southern end of Cceur d’Alene Lake northward to Liard River in British Columbia, about 900 miles, and throughout it has the form of a narrow and remarkably straight depression lying between the easternmost mem- ber of the Rocky Mountains and the rest of the system. This depression is unique among all the mountain valleys of the earth for its remarkable persistence. It is not drained by a single great river, but is occupied in turn by the headwaters of the Flathead, Kootenai, Columbia, Canoe, Frazer, Parsnip, Finlay, and Kachika rivers. The name Rocky Moun- tain trench is now applied to it. The major valley next in order to the west is occupied at the bound- ary by that portion of the Kootenai River that returns into Canada after rounding the great bend at Jennings, Montana. This trench begins on the south near Bonners Ferry, Idaho, and is occupied north of Kootenai Lake by the Duncan River. In line with the Duncan River Valley is the 50-mile trough occupied by Beaver River which enters the Columbia at the Canadian Pacific Railroad crossing. The length of this topographic unit, called the Purcell trench, is about 200 miles. The third major trench is known as the Selkirk trench or the Selkirk 1 Daly, loc. cit., p. 596. 2 F, L. Ransome, Geology and Ore Deposits of Coeur d’Alene District, Idaho, Prof. Paper U.S. Geol. Surv. No. 62, 1900, p. 13. 302 FOREST PHYSIOGRAPHY Valley, and is occupied southward by the Columbia River. Its north- ern extremity is near the 52d parallel, where it is confluent with the Rocky Mountain trench. The southern end of the trench is about 60 miles south of the 49th parallel, where the Columbia River enters the great basalt plateau of Washington. Besides these three first-rank valleys or trenches in the northern Rockies are a number of second-rank longitudinal trenches. Waterton River, Flathead River, and Zigzag River occupy parallel valleys that still further divide the northern Rockies into the Lewis, Livingston, MacDonald, and Galton ranges. Cross trenches or valleys are employed in indicating the minor subdivisions of the principal mountain systems and ranges, as the Slocan Mountains are separated from the rest of the Selkirk System by a depression occupied by Slocan River, Slocan Lake, and the valley of a creek whose mouth is at Nakusp or Arrow Lake.!’ Ina similar way cross valleys divide the Colville Mountains, Pend Oreille Mountains, etc., from the larger ranges. GEOLOGIC FEATURES The detailed topographic qualities of the northern Rockies as dis- cussed in succeeding paragraphs may be better understood by a brief consideration of certain fundamental geologic conditions. From Kootenai River or Purcell trench westward to the Columbia there is a much disturbed, highly folded, rock complex in which the metamor- phism of the sediments generally diminishes_in intensity westwardly.? To this metamorphosed complex the rocks east of the Purcell trench are in striking contrast. They consist of a thick series of arenaceous sedi- ments (pre-Cambrian) which show no pronounced regional metamor- phism. The essential structural features are open folding and exten- sive faulting. They are known to extend from the Belt Mountains westward to the Purcell trench. Some of the ranges (Purcell, Cabinet, Cceur d’Alene, Livingston, and Lewis) are composed almost entirely of this group of rocks. In central Idaho the same strata are invaded, displaced, and probably for the most part cut off by a great granite batholith.? The most striking structural feature of the western part of the sediments east of the Purcell trench is of course the existence of great faults of northwest trend that determine the courses of the trenches and rivers and block out the individual mountain ranges. 1 Daly, loc. cit., p. 590. 2 Ransome and Calkins, Geology and Ore Deposits of the Coeur d’Alene District, Idaho, Prof. Paper U. S. Geol. Surv. No. 62, 1908, p. 16. 3 W. Lindgren, A Geological Reconnaissance across the Bitterroot Range and Clearwater Mountains in Montana and Idaho, Prof. Paper U. S. Geol. Surv. No. 27, 1904, p. 16. | ROCKY MOUNTAINS. I 303 | } Fig. 88. — Coeur d’Alene Mountains, looking from above Wardner. Shows the mature dissection of a ih plateau-like uplift. (Ransome and Calkins, U. S. Geol. Surv.) Fig. 89. — Cceur d’Alene Mountains, looking toward the crest of the range, from valley of South Fork | of Cceur d’Alene River. Shows characteristic equality of ridge lines. (U.S. Geol. Surv.) 304 FOREST PHYSIOGRAPHY CUR D’ALENE RANGE The Coeur d’Alene range, rudely triangular in outline, extends from Lake Pend Oreille on the northwest, southeastward to Lolo Pass (lat. 46° 36’ on the continental divide between Idaho and Montana).! Its southern boundary is vague but should probably be considered as cor- responding to the divide on the northern border of the Clearwater drain- age basin, Fig. 86. Its western margin adjoins the Columbia Plateaus; its eastern boundary is constituted by the valleys of Clark Fork, Flat- head River, and Jocko Creek. The Coeur d’Alene range, Fig. 89, appears as a rather monotonous expanse of ridges nearly equal in height and with somewhat level crests that do not bear prominent summits.” Its broad aspect is that of a maturely dissected plateau, whose general level in its central and highest part is a little above 6000 feet. The rough equality of summit levels becomes less marked toward the west, dissection having there progressed so far that the original regularity of level has now been almost lost. It is tentatively concluded that this topographic uniformity depends upon former base-leveling, but much further work is required for a complete demonstration. The Coeur d’Alene region had been folded and faulted into essentially its present structure and its topography had been developed by probable base-leveling, uplift, and extensive and deep dissection to essentially the present conditions by the time the great Miocene lava flood occurred, for great quantities of basalt flowed up the existing valleys. This fact for example explains Coeur d’Alene Lake, which occupies a large valley that was partly filled with basalt and afterward recut by the original river, thus leaving terraces of basalt on the older slopes. Later a deposit of Pleistogene gravels dammed the valley on the north, originated the lake, and gave it approximately its present outline, backing up the waters of Coeur d’Alene and St. Joseph rivers. PRIEST RIVER RANGE The Priest River range south of the international boundary and east of the Pend Oreille range has been deeply sculptured by glaciers and streams, so that canyons, cliff-bordered cirques, and narrow ridge crests are common. The rocks of the range consist principally of highly fissured granite and syenite; a certain amount of slate and gneiss occurs at the southern end. The average altitude of the range is 5000 to 6000 feet, with a number of elevations attaining 8000 feet. The irregularly fissured condition of the granites results in the better retention of the precipitation than is the case in the schistose rocks of the Pend Oreille range. The latter are either water-tight or else afford too rapid drain- age, depending on the dip of the planes of schistosity. About 91 % of the total forest in the Priest River National Forest is white pine and 1 W. Lindgren, Prof. Paper U. S. Geol. Surv. No. 27, 1904, p. 23. 2 F.C. Calkins, A Geological Reconnaissance in Northern Idaho and Northwestern Mon- tana, Bull. U. S. Geol. Surv. No. 384, 1900, p. 14. ROCKY MOUNTAINS. I 305 tamarack; yellow pine is found below 3500 feet, white pine between 2400 and 4800 feet (best development between 2800 and 3500 feet), and the subalpine fir, of little economic importance, grows above 4800 feet.! CABINET RANGE This range extends from Bonners Ferry southeastward to the junction of Jocko Creek and Flathead River. Its western border is definitely marked by the Purcell trench and its southwestern border by the nearly straight valley occupied chiefly by Clark Fork, Columbia River, and ) Auun Ayadd NN] f the wind gaps are from 5400 to 6000 10nS O the elevati ? 400 feet; Io feet. tends from Mount Heavens north of Mc- x gston range e The Livin Donald Lake northwestward to Mount Head lies west of the Lew It ish Columbia. but the latter range Canada the front range is it in Br the United States ) nm is range mM ends near the international boundary and $99 I ROCKY MOUNTAINS. (‘AINS ‘JOa4} *S ‘Q ‘aJsuvIpenb odAes) ‘ ‘yaaq COS ‘[LAIAJUT INOJUOD ‘eULIWOY ‘S10 [LUOCIYN ayIV] pue slo] ‘Japsoq surejunow jo adAz yovqsoy — ‘£6 “BIy te: = ° oS \ 7 = ° = ‘ fo } W : BE eet WA acon Sid cs Pat G A eet es || “2M uvtueyy” {P3N 0 uN SS. KS p Bi ARR A ~ \ EZ Led ed” [DPQ Tora wa 4919 pau QonB= Bes deeyo SS ; f2ay— . en TIOALZT AGS N “i 0029 2B you £ S300) sulapooy 0081 LOT Wo VTE Ne 310 FOREST PHYSIOGRAPHY the Livingston. The main continental divide follows the Livingston range for some distance southward of the boundary, descending to Flat Top Mountain and finally to the Lewis range, which it follows southward to latitude 46° 45’. Like the Lewis range the Livingston is often nar- row and presents massive mountain peaks of pyramidal outline. Deep valleys diversify its western slope and contain long narrow lakes which vary in length from two to ten miles. The lakes are bordered by slopes of gravel or talus, except at their heads, where cliffs risé precipitously to the range summits. The western limit of the Livingston Moun- tains is definite, but it has the aspect of a bold mountain face rising from foothills rather than the almost sheer face’ of the eastern margin of the Lewis Mountains rising abruptly from the plains. The Lewis and Livingston ranges are characterized by the dominant influence of structure on altitude. In the northern part of the Lewis range and in the Livingston range the greatest altitudes are in gen- eral related to the two anticlines; the master valleys are in the inter- vening syncline, and Flat Top Mountain is the former floor of a broad synclinal valley. A peneplain was formerly developed over the Great Plains and over the Galton range. On the soft rock of the plains it was well developed, but on the harder rocks of the Galton Mountains it was probably imperfect. In the Lewis range it is notable that each peak approaches in height that of its neighbors which stand along the strike in a similar structural position. However in a broad view and taking the Lewis and Livingston ranges as a unit, no general upper limit of heights common to widely distributed peaks may be discerned. If the base-leveled surface was ever in existence in the range, the extreme localized deformation of the mountains has so warped the ancient surface, and intense erosion by both water and ice has so completely dissected it, as to make its determination very difficult if not impos- sible. It is truly remarkable how abruptly the well-developed and but little dissected peneplain of the Great Plains of Montana terminates at the foot of the front ranges. The line is almost as definite as a shore. The long-continued erosion which the peneplain represents must have affected the present mountainous area west of it, but was probably offset by repeated deformation terminating or culminating in the great overthrust to which the present mountain height and the deep dissection are largely due. In the Galton range the old surface may be safely in- ferred; the older features have not been so completely destroyed owing to the relatively small amount of local deformation and the less intense action of water and ice. For the Galton range, although bounded by ROCKY MOUNTAINS. I earat ON —tlow= GY BIN x Icebe G fs GO OZ ms EG I LINN il I Sea rows | U sat! Uae wu Vigginic. 0 SRA \ il am Suga 7a) Wen Lf ALS 30 Scale of Miles 4g 0 1 2 3 4 5 {| Suns perme fear Proper Doro] poe ed LS S07 a 29°35 4 5: Kilometers Contour interval 500 feet Fig. 94. — Map of a part of the Lewis Mountains, western Montana, representing typical glacial features of the range. (Chief Mountain quadrangle, U. S. Geol. Sury.) 312 FOREST PHYSIOGRAPHY structural limits, is internally but a simple uplifted block whose minor flexures or faults are not sufficiently pronounced to interrupt the ancient surface. GLACIAL FORMS Some of the most remarkable erosion features produced by alpine glaciers are to be found in the Lewis range on the continental divide in western Montana. They are represented on the Chief Mountain quad- rangle, Montana, Fig. 94, one of the most interesting topographic sheets yet published, not only because of the exceptional nature of the region but also because of the unusual faithfulness with which the topographer has represented the landscape. The strata of the Lewis Mountains are strongly jointed, a quality which is highly favorable to the plucking action of glacier ice; furthermore they lie nearly flat and boldly overlook the adjacent plains: a set of conditions exceedingly favorable for the development of glacial forms on an unusual scale. Three major topographic features are apparent in the region: (1) numer- ous reversed slopes occupied by lakes and lakelets; (2) strikingly deep wall-like valley sides; and (3) huge amphitheatral valley heads in which a number of living glaciers are to be found. All these features are normally associated with the work of alpine glaciers whose former existence in large numbers has also been determined by the familiar phenomena of eroded and plucked rock surfaces, terminal and lateral moraines, and hanging tributary valleys. Former glaciers plowed down all the main valleys and built up morainic_ accumulations which at Blackfoot and Browning, p. 412, merge into the terminal deposits of one of the great continental ice sheets. An examination of the valley systems in plan shows a remarkable degree of headward cutting on the part of the glacier ice. The amphi- theatral valley heads or alcoves on opposite sides of a divide have in many instances cut back to the point where a knife-like ridge has been formed, as south of Gould Mountain and on the continental divide two miles south of Chaney glacier. The process has resulted in the for- mation of pyramidal mountains such as Going-to-the-Sun Mountain, Cataract Mountain, Little Chief Mountain, Heavens Peak, and others, or in the formation of skeleton mountains of irregular outline such as Almost-a-dog Mountain, Appekunny Mountain, and Merritt Mountain. The manner in which such headward cutting has taken place has been well set forth by Johnson,! who has observed that in glaciated mountains the great curving bergschrund of the snow-field penetrates to the foot 1 D. W. Johnson, The Profile of Maturity in Alpine Glacial Erosion, Jour. Geol., vol. 12, 1904, pp. 560-578. Fig. 95. — Mount Gould, Lewis Range, Montana, looking southwest from South Fork of Swift Current. See Fig. 86 for location. Characteristic cliff of limestone overlooking argillite. The dark band is intrusive diorite. The valley head is a glacial amphitheater developed along joint planes. It is 4670 feet from lake to summit. (Willis.) 313 314 FOREST PHYSIOGRAPHY of the precipitous rock slope constituting the wall of the amphitheater enclosing the snow-filled cirque. He concludes that a causal rela- tion determines the coincidence in the position of the bergschrund and the foot of the cliff wall. The opening allows air to come into contact with both ice and rock at the bottom of the crevasse. By day there is thawing, by night freezing, and blocks of rock are wedged off and the cirque wall riven. The bottom of the crevasse is therefore a “narrow zone of relatively vigorous frost-weathering.” The result is a sapping of the foot of the cirque wall and its gradual steepening and retreat. The continuance of this action steepens the slopes of the glacial amphitheaters and pushes them back until the slopes on the open sides of the divide meet. Thus a more or less rounded divide such as may be found a few miles southwest of Point Mountain, at Flat Top Moun- tain, and other places, was destroyed, the mountain top was reduced to a pinnacle or needle and the divide to a sharp crested aréte. It is this action which appears to have taken place in the Lewis range, and to the varying degrees to which the basal sapping was carried may be attributed the variation of mountain forms ranging between pinnacles on the one hand and flat-topped mountains with small bordering cirques on the other. An extreme case of such extension of headwater amphitheaters may be seen in the Hayden Peak quadrangle representing a portion of the Uinta Mountains of Utah. In this case the amphi- theatral walls have been pushed back to the point where only skeleton ridges remain, and in one instance at least, as west of Hayden Peak, the divide has been almost completely obliterated, so that the snow-fields on opposite sides must have been confluent during the glacial period. It may readily be inferred from these considerations that cirques may be (a) but slightly developed and the present expression of the mountains closely resemble the preglacial expression; or (6) they may beso extensively developed that the upland or mountain region in which they were formed has a fretted appearance; or (c) the cirques may be developed headward to a still greater extent and the dividing ridges trimmed to a row of serrate peaks or skeleton ridges. Further con- trasts in the present characters of the cirques will depend upon the amount of postglacial cutting or filling that has taken place. Most cirques as left by the ice contain but little loose material. Whether or not loose material is now present will depend upon the friability of the rock, the steepness and height of the cirque walls, the precipitation, etc. In the Lewis Mountains the cirques are especially well preserved and still exhibit, Fig. 94, great expanses of little-modified cirque wall, lakes of considerable size and number, and expanses of bare, glaciated rock. ROCKY MOUNTAINS. I 315 Concentrated sapping at the foot of the cirque wall results not only in headward retro- gression but also in downward excavation, and with the melting of the glacier on account of a warmer period the reversed slope at the foot of the cirque wall is occupied by a lake. It will be noted on the Chief Mountain sheet, Fig. 94, that lakes commonly occur in this position. Lakes are also commonly found some distance down valley where tributary glaciers have caused local overdeepening or the bottoms of the glacial channels and the production of reversed slopes back of which the present drainage is impounded. A third group of lakes is frequently found some distance down valley where the drainage has been blocked by morainal accumu- lations that mark the former limit of glacial ice. The valley sides and bottoms in this region usually bear one, two, or more terraces. The lower ones represent a valley filling which the streams are at present cutting away. The higher ones represent inter- rupted valley widening in horizontally bedded rock.! The mountain valleys of these ranges are all noted for their steep walis and lake-dotted and rather flat floors. Their forms are chiefly the result of earlier glacial action. The main glaciers were wide and thick and eroded their channel floors below the channel floors of their tribu- taries. Upon the disappearance of a glacier the former glacial channel became a large part or all of the present valley. Hence the character- istic features of the glaciated valleys and the hanging condition of their tributaries. (See Figs. 96, 97, and 908.) GALLATIN, MADISON, JEFFERSON, AND BRIDGER RANGES In southwestern Montana is a group of mountains whose most important members are the Gallatin, Madison, Jefferson, and Bridger ranges. A small part of the southern end of the Madison range and the greater part of the northern end are composed of gneiss. The central part of the range is an immense laccolith, the uplift due to intrusion having carried portions of the sedimentary rocks almost to the summit of the range and overturned, broken, and changed the strata along the western and northwestern edges of the range, causing a varied and highly irregular topography. The Jefferson range con- sists in the north largely of granitic rocks. The southern section is divided into two parts; the eastern is plateau-like in character, the western is a monocline in which the sedimentary rocks are overturned to greater or lesser degrees from place to place. The Bridger range also consists of three sections, the southern extension being an over- turned monocline of stratified beds that rest upon gneisses which form the western spurs and foothills and also a large part of the main moun- 1 Bailey Willis, Stratigraphy and Structure, Lewis and Livingston Ranges, Montana, Bull. Geol. Soc. Am., vol. 13, 1902, p. 310. 316 FOREST PHYSIOGRAPHY (Davis.) Fig. 97. — The same mountain mass as in Fig. 96, strongly affected_by glaciers which still occupy its valleys. (Davis.) Fig. 98. — The same mountain mass as in Fig. 97, shortly after the glaciers have melted from its valleys. (Davis.) ROCKY MOUNTAINS. I 317 tain mass. The overlying stratified beds form a sharp crest with peaks. The central portion of the range is composed chiefly of sedimentary beds and gneisses are absent. The northern section is also composed of stratified beds of sandstones, shales, and limestones which curve around the ends of the range. The Gallatin range is plateau-like at its summit and is composed largely of volcanic breccias which dip eastward and have their greatest elevations, about 10,000 feet, along the western border. Mount Black- more, one of the most prominent peaks of this range, rises to a height of 10,196 feet. During the general elevation of the region in which these mountains occur the strata were folded and eroded (Cretaceous) and lakes, some of them of great extent, were formed in enclosed fresh-water basins. The lake period lasted for a long time (Neocene to Pleistocene), and during the earlier part of the period there was tremendous volcanic activity. Great quantities of volcanic dust were carried hither and thither by the winds and at length deposited in part on the lake floors as white dust beds; deposition of vast amounts of dust took place upon the adjacent land surfaces, whence the deposits were later washed in large part into the lakes. Later cutting down of the lake outlets allowed these water bodies to become drained and the lake beds themselves to be dissected. The last phases of volcanic activity in the region were flows of basalt and rhyolite, and these now form the summits of mesas, as in the southern part of the Three Forks district. The last geologic episode which has had an influence on the topog- raphy and drainage of the district has been glaciation, but the glaciers were local and the drift deposits in the valleys are of local origin. The low elevation of the ranges, 7000 to 10,000 feet, did not allow vigorous glaciation, and glacial forms have weak expression except in a few favor- able localities. MINOR RANGES OF WESTERN MONTANA In western and southwestern Montana is an extensive area hemmed in by the Madison, Jefferson, and other ranges on the south and the Lewis, Livingston, and other ranges on the north. On the east it extends to the Big Belt Mountains, on the west to the Bitterroots, Fig. 86. The tract includes no prominent mountain chains, only short ranges which reach up to a more or less common level. The geologic conditions are somewhat complex, the rocks consisting of greatly deformed sedimen- 1 A. C. Peale, Three Forks Folio U. S. Geol. Surv. No. 24, 1896, p. 1, col, 4, an Fig. 99. — Effects of slope exposure on forest distribution, western Montana, looking east from Mt. Belmont; see topographic map below. A and C are cool, moist, forest-clad northern exposures, Band D are warm, dry, unforested southern exposures. The trees in the foreground grow on a north- eastern exposure. For the position of these slopes on the map see corresponding letters in Fig. 100. (Barrell, U. S. Geol. Surv.) 7 —Liy LZ) GAFa Z “Jennies CRE \\ = Tak Lay S35 4 \f/ ey HI \\W 0 z ae 2 Miles 1 KB 0 1 2 Kilometers Contour interval 50 feet Fig. 100. — The positions of the letters A, B, C, D, on the map correspond to the positions of the same letters on the photograph. Culture omitted except in case of Marysville. Scale of Miles {a es 218 ROCKY MOUNTAINS. I 319 taries intruded by granite and other igneous rocks. After the deforma- tion of the rocks of the region erosion swept away great quantities of the surface material and reduced the topographic profiles to maturity, Fig. 99. From the standpoint of forestry this great district in western Montana is of special interest because of the strong topographic control of forest distribution, not by control of rainfall distribution but by control of the water supply through variations in slope exposure. Its special signifi- cance may be appreciated by contrast with the physical conditions in the Cascades. The western slopes of the Cascades are wet, the eastern slopes are dry, a rainfall distribution that is directly dependent upon topography and that has a marked effect upon the distribution of the forest trees (p. 164). Among the minor ranges of Montana, on the other hand, there are no great topographic features to obstruct the rainfall and to occasion the climatic contrasts so marked in the Cas- cades. Slope exposure is here the factor of primary importance in forest distribution. The rather evenly distributed rainfall is more quickly evaporated on the sunny southern slopes than on the shady northern slopes, hence the latter are moist and forest covered, the former dry and almost treeless. Since the daily maximum temperature of soil and air is generally attained about two or three o’clock in the afternoon, the driest slope is the one facing southwest. Hence eastern and northeastern slopes are also forested, while western slopes are forest- less. The effect of these conditions is heightened by the action of the prevailing southwest winds which not only dry the southwest slopes but also sweep them clear of snow in winter. The snows accumulate on the northeastern or leeward slopes, where they linger until midsummer and supply the ground and the vegetation it supports with the necessary moisture. These features are exceptionally well developed about Marys- ville and the district therefore merits a somewhat detailed description. In the Marysville district the Rockies are developed in the form of a broad tract extending westward nearly to the Bitterroot Mountains. The eastern portion is a rather flat-topped granite batholith. The entire district has the features characteristic of topographic maturity — “fairly steep slopes, rounded hill crests, and few cliff exposures. On the lower elevations the topography is . . . softened, the view showing successive tiers of well-dissected foothills with slopes of 10° to 20°.” ! The surface is in general covered with a residual soil so thin on the upper slopes as to show a large number of rock outcrops especially on 1 J. Barrell, Geology of the Marysville Mining District, Montana, Prof. Paper U.S. Geol. Surv. No. 57, 1907, D. 4. 320 FOREST PHYSIOGRAPHY the more resistant formations. The lower elevations are covered with a slightly deeper soil, while the main valley floors are deeply filled with alluvium. The Marysville region is in the transition belt between the lower, drier Great Plains on the east and the higher, abundantly watered and forested mountains on the north and west. While the soil covering, though thin, is suitable for a forest growth, the rainfall and snowfall are so light as to support a forest only in favored situations. Marked contrasts in climate occur and are due to marked variations in elevation and exposure. The result is a striking contrast in vegetation on dif- ferent slopes and at different elevations, Fig. 99. “The trees are confined largely to the northern and eastern slopes, since these suffer least from the drying action of the summer sun on the thin soil and also in spring hold the snow longest around the roots. The prevailing winds are from the southwest, with the result that the southwestern slopes are swept more or less bare of snow, which accumulates on the lee- ward side of the hills. Here, protected by the evergreen trees, stray banks linger until about the first of July. “The bottoms of the deeper gorges are especially picturesque, offering the contrast of dark, forested southern! or western walls and grassy northern slopes, while cottonwoods and willows grow in clumps and lines along the courses of the streams. “On the northern hill slopes the trees continue down to elevations of about 5000 feet, the limit varying considerably with the nature of the soil. Below this level the low hills of the northern half of the district are bare of trees and almost without grass, but in the gulches which trench them scattered pines have found enough moisture to give them a foothold. On the lowest levels the prickly pear and bunch grass hold sway. The most desolate portion of the district is north of Little Prickly Pear Creek, for here the sandy, porous nature of the sur- face renders it doubly difficult for vegetation to maintain a foothold, and large areas are covered with nothing but shaly shingle or ancient river cobbles.” 2 MOUNTAINS OF NORTH-CENTRAL IDAHO The mountain ranges of the northern Rockies that we have ex- amined thus far lie chiefly east of the Bitterroot axis. An examination of Fig. 86 will show a vast mountain region north of the Snake River and west of the Bitterroots, a labyrinth of sharp peaks and ridges whose steep slopes descend to deep steep-sided canyons. As a rule the mountains rise suddenly from the bordering plains and ultimately to sharp ridges that attain 11,000 and even 12,000 feet. The origin and nature of a large portion of this wild mountain region may be appreciated best from a view that embraces the contrasting fea- tures of the plains country formed upon the flat basalt sheets that rim about the margins of the mountains and extend far westward into Wash- ington, as far as the eastern wall of the Cascades. The summit of Bald 1 Tt should be remembered that the southern wall of a valley has the same slope exposure as the northern slope of a hill. 2 Tdem, p. 6. ROCKY MOUNTAINS. I 321 Mountain near the common border of the two regions (lat. 46° 25’) affords an extensive view eastward over the level crests and maze Poy 58 a Ve, ~a, = “of Pty, i % oy i) Mp |i) "ey 4 if / Ay gl ss io 2 s 6 x 5 ~ as} oO Sg Q Fig. ror. — View south across Salmon River Canyon, from south slope of Caseknife Mountain, showing plateau char- acter of Salmon River Mountains. (Lindgren, U. S. Geol. Surv.) 8 fc of ridges and canyons that constitute the prin- } cipal features of the Clearwater Mountains (lat. 46°). So thoroughly dissected are these moun- tains that little of their original flat outline may now be seen. For the first 80 miles eastward toward the Bitterroot Valley the lonely trail does not disclose a settlement or even a miner’s cabin.? The irregular canyon courses are the chief routes for transportation by pack mule. It is the worst part of the Kentucky and West Virginia “mountains” set upon the western fringe of the * Probably all granite 10 HORIZONTAL AND VERTICAL SCALE (Lindgren, U. S. Geol. Surv.) The Crags Siese Granite Rockies. ok Four thousand feet below the level of Bald 2 \ Mountain is a scene of far different quality. The lava plain of the Columbia, only gently undulating, stretches out apparently without limit westward. The undulating Camas and Kamiah prairies are \ checkered with waving wheat fields or wild grass, and cultivation and prosperity are brought into j close contact with mountain wilderness. Here we have two regions of great difference in relief but also with many features that indi- River Clearwater Profile across Bitterroot Mountains, Clearwater Mountains, and Columbia Plateau, showing the plateau quality and the rock character. 1 W. Lindgren, A Geological Reconnaissance across the Bitter- root Range and Clearwater Mountains in Montana and Idaho, gs Prof. Paper U. S. Geol. Surv. No. 27, 1904. i Fig. 102. Sea level 0 322 FOREST PHYSIOGRAPHY cate peculiar similarities. The basalt plain is comparatively flat to-day; once it was still flatter; as time goes on it will become more and more irregular, will become indeed very much like the Clearwater Mountains are to-day. The even accordance of hill and ridge top levels in the Clear- water Mountains and the manner in which the plane of the ridge tops cuts across rock of diverse hardness and structure are clear indications that the region was once base-leveled and has since that time been up- lifted and maturely dissected so that flat land is nowhere visible. Dis- section has progressed so far that the once flat tabular summits have been transformed into sharp ridges, but it has not yet progressed far enough for the rivers to have begun to form valley flats. The result is a typical hill- and-valley country, whose resources of forest and mine are difficult of ac- cess, and where agriculture andrelated industries are practically unknown. The sudden descent of the Clearwater Mountains to the level of the lava plain is due not to differential erosion but to pronounced crustal warping or faulting, for there are no structural features that would enable erosion to work out a mountain border of this character. The sudden and notable descent from the level of the plateau to the level of the plain was brought about probably at the time of uplift, for deep canyons were cut following the uplift and before the extrusion of the lava (Miocene) that runs up the old valleys now extending westward under the lavas. A similar abrupt border characterizes the plateau of the Boise Mountains where these descend to the lower valley of the Snake River. About the western base of the Clearwater Mountains the basalt flooded the foothills to a height of 3000 feet and greatly reduced the relief of the region, for the foothills in many cases had a sharply accentuated topography. Above the level to which the lava rose the courses of the streams draining the Clearwater Mountains have been steadily deepening.! Both the Clearwater and the Salmon River mountains are portions of a once gently undulating plateau that has been so deeply eroded by powerful streams that they stand out with great distinctness. The plateau which their crests outline is from tooo to 3000 feet below the summit of the Bitterroot Mountains, causing the latter to appear as a boldly raised block.” Throughout the mountain region of central Idaho the canyons are so deep and steep-sided that many of them are quite impassable.* The 1 W. Lindgren, A Geological Reconnaissance across the Bitterroot Range and Clearwater Mountains in Montana and Idaho, Prof. Paper U. S. Geol. Surv. No. 27, 1904, p. 78. 2 Idem, p. 13. 3G. H. Eldridge, A Geological Reconnaissance across Idaho, 16th Ann. Rept. U. S. Geol. Surv., pt. 2, 1894-95, Pp. 220. ROCKY MOUNTAINS. I 323 canyon walls are precipitous, sheer drops of tooo feet being quite common, _and the adjacent mountain slopes rise by steep ascents 2000 to 3000 feet higher. The most rugged canyons are those of the South Fork of the Boise River, numerous branches of the Clearwater, and the Middle Fork of the Salmon. A stream with typical characteristics is the Salmon River, which heads at the foot of the Bitterroot range, then flows westward through a wide open valley to a point near Shoup, where it enters a profound canyon through which it flows with- out interruption for about 250 miles to its junction with the Snake. The canyon extends through one of the wildest and least-known parts of the state, and is itself so narrow and abrupt and so deep (3000 to 5000 feet) as to be almost untraversable.! The mountainous country north of the Snake River and west of the Bitterroot divide may be divided on the basis of structure into three parts: (1) A great central granite area 100 miles wide and 300 miles long, extending from the Snake River plains northward to an unknown distance but at least as far as 45° 30’; it probably ends near the north- ern border of the Clearwater drainage.’ As thus outlined it forms one of the largest granite batholiths on the continent. Near the lower Salmon River and also near the Seven Devils in the Snake River Valley it is margined by sedimentary rocks. A similar contact occurs on the eastern border of the granite, and from the nature of the intruded beds it is known that the granite mass is probably of Cretaceous age and is an intrusive body similar to the great granitic batholiths of the Sierra Nevada. The granite is remarkably uniform in character except in places upon its margin as on the eastern border of the Bitterroot Mountains where it has a gneissoid structure owing to metamorphism at the time of the formation of the range.? (2) Partly metamorphosed rocks of sedimentary origin— slates and limestones accompanied by schists— occur on the western border of the granite area and form the western border of the mountains. (3) Partly metamorphosed rocks — quartzites, conglomerates, slates, shales, and limestones — occur on the eastern margin of the great granite batholith.* Two types of mountains have been developed upon these rocks and structures: (1) those consisting of masses of rock without any definite range trend and developed upon the almost structureless granite of 1 W. Lindgren, The Gold and Silver Veins of Silver City, De Lamar, and Other Mining Dis- tricts of Idaho, zoth Ann. Rept. U. S. Geol. Surv., pt. 3, 1898-99, p. 78. 2 W. Lindgren, A Geological Reconnaissance across the Bitterroot Range and the Clear- water Mountains in Montana and Idaho, Prof. Paper U. S. Geol. Surv. No. 27, 1904, p. 17. 3 Lindgren and Drake, Silver City Folio U. S. Geol. Surv. No. 104, 1904, p. 1, col. 4. 4 W. Lindgren, The Gold and Silver Veins of Silver City, De Lamar, and Other Mining Dis- tricts in Idaho, 20th Ann. Rept. U. S. Geol. Surv., pt. 3, 1898-99, pp. 79, 86-89. 324 FOREST PHYSIOGRAPHY homogeneous texture. The Clearwater Mountains north of the Salmon River and the Salmon River Mountains south of the Salmon River are the best representatives of this type. The second type of mountain occurs in ranges and is the result of erosion of sedimentary: rock and foliated schists, or a rock complex, whose secondary structures on erosion give a rough trend to the elevated portions. In some of the granites of the western half of Idaho east-northeast- trending structures show chiefly in lines of jointing, in the strike of foliation planes, or in the trend of the fissures. West of the Sawtooth range the divides between the drainage basins are with few exceptions due to early structural features — folds, faults, joints— but over most of the tract the granite is homogeneous, the structural features are only slightly pronounced, the disposition of the topographic elements is a response to the disposition of the early drainage systems, and no well-defined outlines of a range system can be identified.1 The irregular ranges that rise above the general level as in the Sawtooth Mountains seem to owe their existence to greater resistance to erosion rather than to folding and faulting. A few exceptions to this rule are (1) Boise Ridge, which seems to have been partly outlined by orographic disturbances, and (2) smaller ranges which have been developed where the granite is strongly foliated. The mountains of the range type occur chiefly in ihe eastern part of Idaho near the continental divide in a region of pronounced up- lift. They trend in two different directions, east-northeast and west- northwest, depending upon the strike of the beds upon which they are developed. They are composed of altered or unaltered quartzites, schists, and limestones. The Smoky Mountains are an illustration of this type; the Bitterroots are a larger and better known unit and receive more extended description in the succeeding paragraphs. BITTERROOT MOUNTAINS From Hamilton or Missoula in the Bitterroot Valley and in the heart of the northern Rockies one may look west at the bold front of the Bitterroot Mountains. For many miles it maintains a quite remark- able regularity of form and straightness of trend. The slope descends at angles between 18° and 26°. The rectilinear quality is owing to a fault whose locus is the foot of the scarp and whose throw approximates the present difference in elevation between range top and valley bottom. 1 W. Lindgren, The Gold and Silver Veins of Silver City, De Lamar, and Other Mining Districts in Idaho, 20th Ann. Rept. U.S. Geol. Surv., pt. 3, 1898-99, p. 77 325 ROCKY MOUNTAINS. I (‘AIng *Joax *S "f ‘weIspurT) ‘[VAIOJUL JOOF OOF Joe ‘A[UO NOUS YOJYS “euryUOPL puv oCYLp] ‘SUIVJUNOFT 3O0I10}}1g Jo JAvd v Jo deur sIydeIsodoy, —*for “SI STI JO aTB0S WH 7S ‘<2 |\ AA) Z 4 ~~ S = og “I = 2 SAG ey Ze es YUL WS NU 4 ee Z _) oF g = x / WIZE. aes LZ > S pee VA \ a LE —t 04 [TUB FL Wik = | Fy. Z Zyeogol| a) aqaUnas. — LI AGE |. a { Yeo y CES SENSIE'E 326 FOREST PHYSIOGRAPHY At least two groups of facts strongly support this explanation. (1) Faulting occurred in the region so recently as 1898, when for 1500 feet along the base of the mountains a displacement of from 1 to 2 feet was effected and may be observed in favorable localities to-day. (2) All the streams flowing eastward down the regular mountain front have remarkably straight courses, and all have steepened gradients in the last mile or more before debouching upon the valley flat bordering the Bitterroot River. The steepened gradients are an expression of ‘pro- gressive faulting which has prevented the attainment of a profile of equilibrium, approach to such a profile being counteracted by a repeti- tion or continuance of faulting.! The remarkable regularity of the eastern slope of the Bitterroot Mountains is the more interesting because of the preservation of the fault plane throughout the slope up to the eastern summit, a fact which appears to be explained by the flat angle at which the slope was formed, the uniform character of the crystalline rock (gneiss and schist) com- posing the eastern front of the range, and the original regularity of the fault. The main divide of the Bitterroots is a succession of sharp craggy peaks alternating with deep saddles at the heads of the large canyons where glacial cirques have opened up the valley heads. The western slope is more rugged than the eastern, although even the latter is almost unknown except to prospectors. The immense steep-sided gorges, sheer precipices, and extensive rock slides make the western slope en- tirely impassable toward the crest.?. The trails, originally located by the Indians, follow the divides or primary ridges as closely as possible, tortuous as they are; only grading would make the canyons passable. CLIMATIC FEATURES; VEGETATION The character of the primeval forest of northern Idaho and north- western Montana varies according to latitude, altitude, and conditions of exposure. The size of the trees decreases toward the north; on the slopes of the Coeur d’Alene Mountains it is not uncommon to find trees 2 or 3 feet in diameter, while tamarack, spruce, and lodgepole pine along the international boundary are seldom more than 1 foot in diameter. The valley terraces along the principal rivers support large groves of yellow pine and tamarack, almost without underbrush. The moun- 1 W. Lindgren, A Geological Reconnaissance across the Bitterroot Range and Clearwater Mountains in Montana and Idaho, Prof. Paper U. S. Geol. Surv. No. 27, 1904, p. 40. 2 J. B. Leiberg, Bitterroot Forest Reserve, 20th Ann. Rept. U. S. Geol. Surv., pt. 5, 1900, p. 319. ROCKY MOUNTAINS. I 327 tain slopes have a denser cover of fir, spruce, and tamarack, which with increasing elevation becomes an open growth of spruce, with a luxuriant cover of grass. Snow does not persist throughout the year upon any of the slopes much exposed to the direct rays of the sun, though perennial banks form on the steep northward-facing sides of the higher peaks and one or two small glaciers occur in the Cabinet range. There is a wet and a dry season, the latter beginning about October 1, the former about June 15, and the summers are very mild and agreeable. The mountainous part of northern Idaho and north- western Montana has a thin population engaged in lumbering, agri- culture, and mining.' The vegetation of the mountainous area immediately north of the Snake River plains consists of forests of fir and pine, above an eleva- tion of 5000 feet and gradually increasing in luxuriance northward. The southern foothills of the main mountain area below 5000 feet are barren. The agricultural population of the Snake River Valley is con- centrated chiefly where the tributaries issue from the mountains and on the limited flats along the main river, where irrigation is possible. Besides these plains settlements are others in the intermontane valleys on the Weiser, Payette, and upper Salmon. Scattered mining settlements are also found from south of the mountains to Florence, and from the Seven Devils to Challis. They are often in places very difficult of access and at elevations ranging from 4000 to 8000 feet. In the Bitterroots the soils of the mountain slopes are composed of granite débris below and loam above; in the canyon bottoms similar conditions obtain, but the top layer is usually heavier; in the subalpine meadows the subsoil is a pure granite gravel and the surface layer a loam varying from 6 inches to 6 feet in depth.’ Forests of economic value are found in the Bitterroot Mountains, in the upper valleys of the branches of the Bitterroot River, in the can- yons of its tributaries farther north, and on the lower slopes of the mountains. At greater altitudes and upon the sides and summits of the mountain spurs the forests are thin and of little value. Two zones of forest distribution may be distinguished, the dividing line lying about 5800 feet above sea level. The lower is the yellow-pine zone, the upper the alpine-fir zone. The timber of the lower zone consists 1 F. C. Calkins, A Geological Reconnaissance in Northern Idaho and Northwestern Mon- tana, Bull. U. S. Geol. Surv. No. 384, 1909, pp. 20-21. 2 J.B. Leiberg, Bitterroot Forest Reserve, roth Ann. Rept. U.S. Geol. Surv., pt. 5, 1897-98, Pp. 262-267. 328 FOREST PHYSIOGRAPHY mainly of red fir and yellow pine in the proportions of 60% and 30%; in the subalpine zone nine-tenths of the timber consists of lodgepole pine of little commercial value.' The former type of growth prevails on the lower slopes and in the canyons; the latter occurs on the summits and ridges and on the steep upper slopes. 1 Henry Gannett, roth Ann. Rept. U. S. Geol. Surv., 1897-08, p. 57. CHAPTER XIX ROCKY MOUNTAINS. II CENTRAL ROCKIES BETWEEN the northern and southern Rockies is a group of ranges whose principal members are the Absaroka, Wind River, Gros Ventre, Teton, Laramie, and Medicine Bow mountains. Part of them form a belt of broken, rugged country with alpine characteristics on the western border of Wyoming, such as the Teton and Gros Ventre ranges; another part, including the Wind River and Laramie ranges, extends eastward and southward, making a great curve through central and southeastern Wyoming in continuation of the Colorado Range of central Colorado. All these ranges (and many others of lesser extent not described here) are sufficiently alike in certain general characteristics and in geo- graphic position to form a family of ranges with prominent traits. Many of them are anticlinal in structure, the anticlinal uplift being directly related to the mountainous relief, so that uplift along axial lines is responsible for the trend of the range heights. The asymmetry of the folds is another group quality and in the anticlinal ranges always causes one mountain flank to be relatively less steep than the other. Erosion is of course variable in degree, because the degree of folding and the height of the folded strata are not everywhere the same nor are the resistances of the different strata uniform. It is sometimes ventured that these structures and forms resemble those of the long, narrow, Appalachian ridges of Pennsylvania. The comparison is not good as regards structure, for the Wyoming ranges have a far less regular structure and trend than the Pennsylvania ridges. In regard to form the comparison is wholly at fault. No central core or axis of crystalline rock is found in the latter case, and no such heights are reached. The Pennsylvania mountains are ridges; the Wyoming mountains are ranges. In the one case the even sky line is the result of base-leveling; base-leveling has not been generally determined in the other case and the sky line is serrate, the crests and peaks rugged and in some places lofty. Some of the foothill ridges developed upon the 329 330 FOREST PHYSIOGRAPHY Plateau é dison srs Ma = zpund— = Mes, | Rae e Platte R. + On, al a, encetwate, Plat. = Se 2a Bo 5 28 38 eS Grant, Mae. Sweetwater Te $ OF Sweetwater Mts, ‘Semino Mts- Great Divide Basin i Sw MW RY 41uas3q aay 109 Fig. 104. — Outline map of the central Rockies, showing location of principal ranges. edges of flanking strata are nearly level for miles, have subsequent valleys developed upon soft strata alternating with hard, and in general show a rough resemblance to Appalachian mountain topography; but on the whole there are far more contrasts than resemblances. LARAMIE MOUNTAINS The great Colorado or so-called Front Range terminates near the Colo- rado-Wyoming line and the mountain axis of which it is but a part continues northward in a douvle line of uplifts known as the Laramie Mountains on the east and the Medicine Bow Mountains on the west. The southern end of the Laramie Mountains is near the North Fork of the Cache la Poudre River and is surmounted by a confused mass of peaks of considerable height. The front range is also elevated north of the Laramie River, where it rises in rugged granite hills. But between the North Fork of the Cache la Poudre on the south and the Laramie ROCKY MOUNTAINS. II 331 River on the north is an 80-mile division of the front range! of the Rockies whose topographic features are in strong contrast to the rugged groups both north and south. The average elevation of this section is only 7800 to 8300 feet and all but one of the few isolated peaks that rise above the general level are under gooo feet above the sea. Sanders Peak, 9077 feet, Central Peak, 8744 feet, and Arrow Peak, 8683 feet, are the principal eminences. The so-called Laramie Mountains are really an elevated plateau some 1500 feet above the Laramie Plains on the west and somewhat more above the Great Plains that lie beyond their eastern foothill ridges. The plateau is cut by many canyons and roughened somewhat by nu- merous knobs and short ridges, but these features do not wholly obscure the expression of the uplifted peneplain (early Tertiary) formed alike upon granite and schist. The interstream areas are smooth or gently _ rolling and even the ridges have rounded summits. The latter together with the knobs are residual with respect to the peneplain. The rocks are deeply decayed and thick residual soil mantles the remnants of the ancient surface. The main divide of the Laramie Mountains is a distinct ridge of sand- stone and limestone bordered on the east by a scarp from 50 to 200 feet high; on the west this ridge descends regularly to the Laramie Plains as a dip slope. East of the main divide is a broad area of gneiss and schist, the core of the anticlinal axis. Farther east are the hogback ridges of sandstone and limestone that form the eastern foothills. The Laramie Mountains are structurally a great anticline whose summit has been worn away. To the fact that it is an asymmetrical anticline is due the gentle western and the steep eastern slopes of the mountains. Locally, faulting has occurred on the eastern border and heightened these topo- graphic contrasts. At other places the contrasts are lessened by an extensive overlap of Tertiary deposits which on the east extend far up the mountain slopes concealing the underlying harder rocks and afford- ing easy ascents to the mountain or plateau summit.? There is a noticeable lack of timber over this entire section. Small groves of pine, aspen, and other types of tree growth are found in sheltered localities, as in the larger basins, but on the whole the timber is of little practical importance.’ 1 The name applied to the easternmost great range of the Rocky Mountain System. In Colorado the front range is the Colorado Range; in Wyoming it is the Laramie and other mountains; in Montana it is the Lewis Mountains, etc. 2 Darton, Blackwelder, and Siebenthal, Laramie-Sherman Folio, U.S. Geol. Surv. No. 173, IQIO, PP. I, 2, 13. 3 Arnold Hague, U.S. Geol. Expl. of the 40th Parallel (King Surveys), vol. 2, 1877, pp. 5-7. 332 FOREST PHYSIOGRAPHY WIND RIVER RANGE The Wind River Range trends southeast and is one of the eastern border ranges of the central Rockies, Fig. 104. It continues toward the north in line with the Laramie Range, and like it is one of the front ranges of the system. The general structure of the Wind River Range is that of a complex anticlinal whose short and steep dips and corres- ponding steep slopes are toward the southwest; the northeast descents are less steep. The latter aspect is varied by two parallel mountain ridges formed upon hard strata, the intervening soft strata having been cut away by strike or subsequent streams. The main summit of the mountains consists of a broad plateau-like tract of granite and gneiss whose northeastern and southwestern faces are En eaters and deeply scored by large canyons.! In the Wind River Range spruce and fir constitute the main tree species, with occasional groves of yellow pine and quaking aspen. Below the foothill elevations the willow and the quaking aspen grow along stream courses; at still lower elevations and in dry interstream situa- tions sage-brush and cacti are the characteristic plants, as in the low country east of the Wind River Range and north of the Sweetwater Plateau. On dry ridges and steep dry bluffs pifon and cedar, or more properly, juniper, are found.” BEARTOOTH AND NEIGHBORING PLATEAUS The Yellowstone River after leaving Yellowstone Park makes a great northerly bend between which and the Bighorn basin are the Bear- tooth Mountains or Plateaus, East and West Boulder plateaus, Lake Plateau, etc., and the northern end of the Absaroka Mountains. In fact the three last-named tracts, together with the Buffalo Plateau and other minor subdivisions, constitute the northern part of the Absarokas as shown in Fig. 05. They are drained northeasterly by the Stillwater, Clarke Fork, and other tributaries of the Yellowstone, through deep canyons bordered by wall-like cliffs from 1500 to 3000 feet high. The region is everywhere extremely rough, and scored by deep, narrow, rocky canyons between which are (1) narrow ridges often only a few feet wide at the top or (2) broad, massive, rolling, bowlder-strewn, plateau-like surfaces like the East and West Boulder plateaus. The 1 Orestes St. John, U. S. Geol. and Geog. Surv. of the Terr. (Hayden Surveys), 1877, pp. 228 et seq. '2¥F. M. Endlich, U. S. Geol. and Geog. Surv. of the Terr. (Hayden Surveys), 1877, pp. 59-50. ROCKY MOUNTAINS. II 333 BS ) a Lf 9) ) c ( IW (O00 | PUN Pies (URE g Xs Y | | « 00 ith Jas NN nN } Ww Lue y QO yy : SS \\ Mie gz NY S\\\ EIS re I\\ Y. ys Matava Y ES ) <« Ss LA K Ls Mh N ey) UGS 10}15 in b-°Yp Nt Si SNS Fig. 105. — Topography of East and West Boulder plateaus, northern end of the aroka Range, sout western Montana. (Part of Livingston quadra S 8 10 Contour interyal 200 feet 334 FOREST PHYSIOGRAPHY average elevation is about 8000 feet; the canyon floors are at 4000 and the highest peaks at 11,000 feet. The Beartooth Plateau, the main mass of high, rolling, inter-canyon country, and the associated ranges or plateaus, lie east of the Stillwater and at higher levels, g000 feet, with peaks reaching to 13,000 feet. The canyons are here also steep and deep and end in rock-bound glacial cirques whose development about the bases of the mountains has sharpened the latter from a more rounded form into peaks and pinnacles of the matterhorn type that give an aspect of profound relief. The main plateau is pitted with glacial depressions, many of which have lakes and marshes bordered by alpine meadows. ‘On the western border of Beartooth Plateau is the deep canyon of the Stillwater; on the east the plateau ends in a great frontal scarp in some places more than 3000 feet high. The steepness of most of the forms is owing to the late uplift of the region, the profound dissection by both water and ice, and the resistant gneisses and other crystallines out of which they have been carved. Some of the canyons are mere rifts between almost perpendic- ular walls; most of the headwater cirques are sheer cliffs; the peaks are generally glaciated and their bases carved by ice and water to such an extent as to give them great steepness and sharpness of form both in general outline and in detail. The plateaus and canyons in the entire region as well as in the Absa- roka and Gallatin ranges on the west were entirely covered with snow and ice during the glacial period; in addition all were profoundly dis- sected even before glaciation set in. Each range therefore repeats the general features of its neighbor. All are characterized by broader or narrower plateau-like tracts between deep canyons; all have some sur- mounting peaks; all have glacial lakes (1) on the cirque floors, (2) on the plateau summits, and (3) in the valleys, where either unequal cutting or morainic damming originated basins of variable size. The forests of the region are almost wholly coniferous. Limber pine grows in belts from 5500 to 6000 feet high; lodgepole pine grows at elevations ranging from 5000 to 8000 feet; red fir is most commonly found on dry rocky slopes; and Engelmann spruce grows along the canyon bottoms or along seepage lines. Above the lodgepole pine forest are subalpine fir, white-bark pine, and Engelmann spruce. Timber line is at 9300 feet on northern and western slopes and at 9800 feet on south- ern slopes. ‘Toward the east it rises and on the eastern edge of the Bear- tooth Plateau is at 11,000 feet. At the upper limit of tree growth the spruce and the white-bark pine decrease in height and become mere crooked shrubs. Grass grows in the alpine zone in such abundance as ROCKY MOUNTAINS. II 335 to form a thick turf capable of storing water and adding to the effect of lakes and tarns in preventing too rapid run-off and gullying. Overgrazing immediately starts gully erosion and results in effects as important as the overcutting of the forests at lower elevations.! ABSAROKA RANGE The Absaroka Range extends in a north-south direction for over 80 miles, with an average width of nearly 50 miles, Fig. 105. The southern end of the range is closely related to the Wind River Plateau and the Owl Mountains and is made up of enormous volcanic flows. The Absa- roka mountains are to be considered as a broad deeply-eroded plateau rather than a sharply outlined range, Fig. 105. Profound erosion has carved the former more extensive summit into a topographic complex of rugged peaks and jagged pinnacled mountains bordered by bold escarp- ments which rise to heights of hundreds and even thousands of feet above the surrounding country. The walls encircling the mountain groups are usually abrupt and owe much of their steepness to ice action. After the volcanic materials of which the Absaroka mountains are largely composed had accumulated the streams cut the range into isolated mountains and broadly-spreading spurs and its deep broad valleys were occupied by glaciers. All the upper valleys on the eastern slopes of the mountains have been sculptured by ice action, and lateral and terminal moraines are abundant. In addition, many narrow can- yons are broadly benched a thousand feet or more above the valley floors, and rounded and polished forms are characteristic features of the scenery wherever gneisses and granites occur. The chief effects of glaciation are, however, to be found upon the west- ern slopes of the Absaroka Range, where a heavy ice cap developed, — one of the largest local glacial centers developed in the Rocky Mountains south of the great continental ice sheet. This local ice cap was confluent with that in the Yellowstone National Park, where a broad area of elevated country supplied favorable conditions for extensive glacial accumulations which fed marginal glaciers deploying down the valleys.? In the deep basins of some of the higher valleys facing northeast, a few snow-fields are still to be found and even small glaciers. The glaciers lie in broad, 1 J. B. Leiberg, Forest Conditions in the Absaroka Division of the Yellowstone Forest Reserve, Montana, and the Livingston and Big Timber-Quadrangles, Prof. Paper U. S. Geol. Surv. No. 29, 1904, pp. 10-21. 2 W. H. Weed, The Glaciation of the Yellowstone Valley North of the Park, Bull. U. S. Geol. Surv. No. 104, 1893, p. 13. 236 FOREST PHYSIOGRAPHY interval, 100 feet. (Ishawooa quadrangle, U. S. Geol. Surv.) AS ANS a MSs VY HHA e : | & psy i | (( \ / Fla WE ame Wa \\\ | Fig. 106.— Topography of the Absaroka Range 10 miles east of Yellowstone National Park. Scale, 234 miles to the inch; contour ROCKY MOUNTAINS. II 337 rock-bound amphitheaters in great measure protected from the direct rays of the sun between high walls, and in localities where the prevailing southwest winds deliver vast quantities of snow during the winter season.! The greater part of the Absaroka Range in the Yellowstone National Forest is clothed with coniferous forests broken by open glades. The isolated peaks and irregular crests of the main ridges are above timber line, and bear only scattered and stunted growths of weather-beaten trees. The western side of the range has a more continuous forest cover than the eastern. Lodgepole pine is the prevailing tree, limber pine is found at higher altitudes, and balsam and spruce are scattered widely though they nowhere attain great height or size. The most stately and vigorous tree of the Absarokas is the Douglas spruce, but it has a scattered growth. None of the timber of the Absaroka region is of superior quality, though sufficient for local requirements may be obtained by judicious use of the forest.’ MEDICINE BOW RANGE This range is about 100 miles long and diverges slightly with respect to the front range. Included between it: and the front range is a high intermont basin, so-called Laramie Plains. Its broadest part, in the region of Medicine Peak, is from 30 to 35 miles across, but the southern end is only 10 or 12 miles wide. The highest peaks named from the southern end northward are Mount Richthofen, 13,000 feet; Clark’s Peak, 13,100 feet; Medicine Peak, 12,200 feet; and Elk Moun- tain, 11,500 feet. North of the 41st parallel the range becomes double crested in response to the double anticlinal structure, but toward the south its unity is preserved. Between the double crests of the northern section is a high, gently undulating, intermont plateau 10,000 feet above the sea, whose surface is covered with timber and dotted with lakes. All the higher portions of the range have been glaciated, and some of the glacially carved amphitheaters are very striking, with steep walls 1500 feet high cut in extremely hard quartzite. The greater part of the range is covered with coniferous forest which is in places quite dense. Douglas spruce, Engelmann spruce, and yellow pine are the chief species. The cold timber line is about 11,000 feet above sea level.’ 1 Arnold Hague, Absaroka Folio U. S. Geol. Surv. No. 52, 1890, p. 6, col. 3. 2 Idem. 3 Arnold Hague, U. S. Geol. Expl. of the 4oth Par. (King Surveys), vol. 2, 1877, pp. 94-07. 338 FOREST PHYSIOGRAPHY BASIN PLAINS OF SOUTHERN WYOMING AND MINOR RANGES ON THEM An accurate relief map of the Rocky Mountains represents a broad and flat tableland between the Laramie Mountains and the ranges west of the Green River. Through the eastern portion of this tract runs the Medicine Bow Range and through the central portion run the Leucite Hills, etc., so that the tract is partially divided into three sections. The portion east of Medicine Bow Range is called the Laramie Plains; Fig. 107. — Laramie Plains, looking southwest from near Mandel, Wyo. (U.S. Geol. Surv.) the central section is known as the Red Desert, etc.; and the western- most section is known as the Green River Basin. These three main divisions are, however, continuous about the northern end of their moun- tainous boundaries. The Laramie Plains are 7000 feet above sea level, have a relatively smooth and gently undulating surface, Fig. 107, with gentle slopes and rounded outlines, so that they appear practically level over broad ex- panses except where a few bench-like ridges or very low buttes of cir- cumdenudation occur. Shallow lakes, none more than a few square miles in extent, are scattered over the surface of the plain, whose ROCKY MOUNTAINS. II 330 waters range from fresh to brackish or strongly alkaline; some of them disappear during the dry season and leave saline incrustations. The whole tract is a great natural pasture; trees grow only along the broa stream valleys.! The Laramie Plains apparently owe their flatness to the general horizontal attitude of the underlying strata (Cretaceous). The highest observed dips in the central portions of the basin are 12°, the average 5° to 8°. On the borders of the basin the dips increase somewhat and the eroded strata present steep bluffs toward the older rocks that form the cores of the bordering ranges. One of the most typical portions of the high intermont plains of southern Wyoming is the central division, which extends west of Rawlings. There are but few exposures of the flat-lying beds, and these occur only in escarpments a few feet high. The barren surface is but slightly dissected by the dry, shallow water- courses that traverse it. It is a flat, monotonous region as far as the Leucite Hills and is broken only at long intervals (1) by hills of erup- tive material, or (2) by lines of moving sand dunes and drifts, as west of the Red Desert and north of the Leucite Hills, or (3) by local uplifts of small extent as at Rawlings. No marked drainage features are found in this section, but the section west of the 1ogth meridian is drained by the upper tributaries of the Green and along their courses an im- portant amount of dissection has taken place.’ The westernmost section of the Wyoming Plateau is also under- laid by nearly horizontal strata (Tertiary and Cretaceous). The general nature of the relief is similar in most respects to that of the eastern divisions. The central depression of the tract is occupied by the valley of Green River, which on the north flows through the basin in a wide alluvial bottom; farther south it flows through a 1ooo-foot canyon cut in nearly horizontal strata (Tertiary). It then enters the Uinta range at the Flaming Gorge, where it cuts through the hardest quartzite, turns out of the mountains, flows eastward, and finally cuts straight across the Uintas in a superb 3000-foot quartzite canyon known as the canyon or gate of Ladore.? — The Green River drains a great basin whose eastern portion is some- what diversified by low, strike ridges of irregular outline and whose western portion bears ridges of similar structural nature but of more regular topographic development, the horizon being always bounded 1 Arnold Hague, Rocky Mountains, U.S. Geol. Expl. of the 40th Par. (King Surveys), vol. 2, 1877, PP. 73-75- 2 S. F. Emmons, The Green River Basin, U.S. Geol. Expl. of the 4oth Par. (King Surveys), vol. 2, 1877, p. 164; and Geol. Map by Peale, St. John, and Endlich, accompanying Report. 3 Idem, pp. 192-193 and 287, with Plate 8. 340 FOREST PHYSIOGRAPHY by an almost perfectly horizontal line.| The southern portion of the region north of the Park Range, known as the Savory Plateau, likewise presents a horizontal sky line, but for a different reason. The surface strata (Tertiary) are horizontal here and overlie the upturned and eroded edges of Cretaceous and older beds and present smooth plain surfaces.” From these stratigraphic relations it is reasonable to con- clude that a mountain-making period at the close of the Cretaceous or — Lae ey as Fig. 108. — Terrace and escarpment topography of nearly horizontal beds (Eocene), Green River Basin, southwestern Wyoming. (Veatch, U.S. Geol. Surv.) in early Tertiary deformed the underlying beds and erosion planed them off, some perfectly, others imperfectly, and that upon the subdued, partially reduced surface thus developed, Tertiary beds of great thickness (up to rooo feet) were deposited. Erosion dissected (1) the Cretaceous strata that were never buried, causing the development of a ridge and valley type of topography, and (2) the Tertiary cover, in some 1 §, F. Emmons, The Green River Basin, U. S. Geol. Expl. of the 40th Par. (King Sur- veys), vol. 2, 1877, p. 192. 2 Idem, p. 164. ROCKY MOUNTAINS. II 341 instances, laying bare the Cretaceous beneath, in others exposing only Tertiary material in the valley sections. Similar relations have been identified in southwestern Wyoming and on the southwestern border of the Green River Basin. The basin floor is here developed chiefly upon Tertiary strata dipping gently eastward, and in general has been eroded so that the strata present their outcrop- ping edges as westward-facing escarpments, a typical terrace topog- raphy. In places the mantle of nearly horizontal Tertiary deposits has been worn away and the underlying older beds with steep inclina- tion are now exposed. Since there is considerable difference in the exact degree of dissection of the capping strata and of the exposed portions of the underlying, deformed rocks, the topography also varies Fig. 109. — Hogback topography of inclined beds (pre-Eocene), southwestern Wyoming. The crests of the ridges are developed upon sandstone underlain by shale. (Veatch, U.S. Geol. Surv.) from place to place. Two main types of topography have been de- veloped: (1) flat table-like forms in places with bordering escarpments of considerable length between benches that rise in regular succession, a topographic type developed upon the uppermost and horizontal beds; (2) long ridges, often sharp, separated by equally long valleys, the whole in a markedly parallel arrangement and developed upon the steeply in- clined older rock composed of alternating hard and soft layers. Hard 342 FOREST PHYSIOGRAPHY sandstones are the ridge makers; soft shales underlie the valleys and the lowlands.1 The greater portion of the Green River Basin presents a gravel-strewn surface almost everywhere in process of dissection. The gravel cap was formed (1) by the weathering in place of the partly consolidated or wholly unconsolidated deposits filling the basin, or (2) by detrital accu- mulations washed into the basin during an earlier period of aggradation. The latter material is abundant about the mountainous border of the basin and originated in the form of desert-fan deposits. Its origin may be related to the renewed growth of the bordering ranges. Its deposi- tion was preceded by a long erosion interval in which a local peneplain was formed, a fact which partly explains the extent of the gravel cap. It was followed by erosion now in progress.” The Leucite Hills, which partially break the continuity of the plains of southern Wyoming, consist of a number of small conical peaks of volcanic origin. Some of them are capped by lava flows, others have crater-like forms and appear to have been centers of eruption.? They have no great topographic prominence and are of little interest in the present connection. SIERRA MADRE RANGE Continuing with the mountain ranges of the central Rockies it should be noted that in south-central Wyoming (Encampment District, south of Rawlings and west of the Medicine Bow Range) the mountain topog- raphy is on a whole of a subdued type. Steep slopes are confined to the middle courses of a few of the main streams and to a few basins or amphitheaters near the main divide formerly occupied by glaciers. The Sierra Madre mountains here form the main or continental divide. Their summits are generally broad and form a flat surface whose eleva- tion ranges from 9200 feet in the lowest pass to 11,000 feet on the high- est summit. The steep headwater declivities on opposite sides of the divide are often a mile or more apart, and the soft, broader-spaced con- tours of the mountain summits are frequently continued out upon the marginal spurs for several miles. The high portion of the range, Fig. 110, thus appears as an elevated plateau submaturely dissected by existing streams. The topography is subdued rather than rugged, and wagon 1 A. C. Veatch, Geography and Geology of a Portion of Southwestern Wyoming, Prof. Paper U.S. Geol. Surv. No. 56, 1907, pp. 34-35, and Plate r. 2 J. L. Rich, The Physiography of the Bishop Conglomerate, Southwestern Wyoming, Jour. Geol., vol. 18, 1910, pp. 601-632. 3S. F. Emmons, U. S. Geol. Expl. of the 4oth Par. (King Surveys), vol. 2, 1877, p. 236. ROCKY MOUNTAINS. II 343 roads have been constructed on almost every portion of the area with- out that excessive expense that is usually required in opening trans- portation routes in mountainous regions.! The general structure of the Sierra Madre Mountains is a low arch or anticline the axis of which is parallel with the mountain crest. This arch was gradually uplifted and eroded and the Mesozoic rocks removed from the axial portion, revealing older pre-Cambrian rock which forms the main mass of the mountains. The Mesozoic formations outcrop on the foothills on either side and dip away beneath the surrounding plains. been? SAS SSR dN Wie ty RSI, pO Y 2) 398 , AY, WZVASG Fig. 110. — Mature profiles, long gentle spurs, and rounded summits of the mountains of the Encampment district, south-central Wyoming. (Encampment Special quadrangle, U. S. Geol. Surv.) TETON RANGE The Teton Range extends almost due south from the southwestern corner of Yellowstone Park and just east of the Idaho-Wyoming line. It is about 40 miles long and from 1o to 15 miles wide, with a central cluster of exceptionally high peaks, the highest of which, Mount Hayden and Grand Teton, attain elevations of 13,700 feet and 13,800 feet re- spectively. The only practicable pass is Teton Pass at 8400 feet. There are still in existence several small glaciers occupying cirques high 1 A. C. Spenser, The Copper Deposits of the Encampment District, Wyoming, Prof. Paper U.S. Geol. Surv. No. 26, 1904, pp. 12-15. 344 FOREST PHYSIOGRAPHY up in the range. The Tetons consist of a great longitudinal axis of uplift and folding; the steep and short descent is on the east, the longer and gentler is on the west. The steep eastern front is cut into a number of great buttress-like spurs faced by bold precipitous cliffs. On the west are a number of long narrow canyons separated by broad westward- descending slopes. These features are very uniformly developed along the entire eastern and western mountain fronts and are the most im- portant in the range. The main mountain forms have a very intimate relation to a single great anticlinal uplift greatly eroded, exposing a central core or nucleus of crystalline rock, chiefly gneiss and granite, the latter forming the sharp peaks and aiguilles as well as the highest peaks as in Mount Hayden; the gneiss tends to erode into the form of sharp ridges. The crest of the range does not everywhere follow the granite however. At the south it is developed chiefly upon sedimentary rocks that have not yet been stripped from the crystalline foundation.! In general the climatic conditions in the Tetons favor forest growth, yet but little forest exists owing altogether to the repeated and destruc- tive fires which have swept over the tract. The lodgepole pine has been - temporarily driven out in many places and former timbered areas have become grass-covered parks or aspen groves. Besides lodgepole pine and aspen the principal species are Engelmann spruce and red fir, the former in damp, the latter in dry situations. Timber line is at 10,000 feet in this locality, and as the average altitude of the range is about 12,000 feet, large portions of the Tetons rise well above the upper limit of the forest. GROS VENTRE RANGE The Gros Ventre Range consists of two parallel mountain folds or anticlines about 5 miles apart, the axes trending southeast. Erosion has unroofed both anticlines and given them corresponding topo- graphic features. Had the anticlines been symmetric, long, gentle, dip slopes would, in the present state of erosion, be found upon the outer flanks of the ridges, and steep, short slopes would be found cutting across the inner edges of the strata. The folds are, however, not sym- metric, for the axes lie near the southwestern margins, hence the south- western slopes are steep, the northeastern slopes relatively gentle and uniform in character. ‘The contrast is identical in general kind, though 1 Orestes St. John, U. S. Geol. and Geog: Surv. of the Terr. (Hayden Surveys), 1877, Pp. 411-4106. 2 T. S. Brandegee, Teton Forest Reserve, roth Ann. Rept. U. S. Geol. Surv., pt, 5, 1897-08 Pp. 195-107. ROCKY MOUNTAINS. II 345 somewhat dissimilar in detail to that afforded by the eastern steep and western less steep slopes of the Tetons. As in the latter case also, erosion has denuded the sedimentary cover and exposed portions of the underlying crystallines. These do not, however, form any notable por- tion of the mountain crest, for denudation is here far less advanced and the mountain summits are still largely developed upon sedimentary rock.! EXTRA-MARGINAL RANGES On the eastern and the western borders of the central Rockies are two mountain ranges which stand out prominently at a little distance for- ward from the main mountain front. The eastern range is the Bighorn Mountains; the western range is the Uinta Mountains. Although they are here classified as a portion of the central section of the Rockies they should be regarded as having no very close geologic or geographic affin- ities with the Wyoming ranges described above. UINTA MOUNTAINS The Uinta range is peculiar in trending in an east-west direction in a mountain region where the prevailing trends are north-south. It is not altogether unique in this respect, however, for the Owl Creek range is a fold having a similar trend, and others in Wyoming and elsewhere hav- ing the same trend have been discovered.” The Uintas form a rather flat, elliptical dome or elongated arch 150 miles long from east to west and with an average width of 20 to 25 miles. The interior of this elongated dome is a deeply dissected, plateau- like region, in general about 10,000 feet high, from which rise narrow ridges and peaks 12,000 to 13,000 feet high developed upon horizontally bedded quartzites.® The strata of the interior plateau nowhere depart more than 5° or 6° from a horizontal position except in a small number of local instances. On the margins of the mountain belt the beds dip more steeply, in some cases over 45° and in extreme instances nearly go°. The marginal zones of highly inclined strata are also affected by minor faults and folds, the principal displacement being along the northern side of the arch, Hig. 11 1.* 1 Orestes St. John, U. S. Geol. and Geog. Surv. of the Terr. (Hayden Surveys), 1877, Pp. 208 et seq. 2 N. H. Darton, Senate Document No. 219, 1906. 3S. F. Emmons, Uinta Mountains, Bull. Geol. Soc. Am., vol. 18, 1907, pp. 287-302. 4 W. W. Atwood, Glaciation of the Uinta and Wasatch Mountains, Prof. Paper U. S. Geol. Surv. No. 61, 1909, p. 9. See also J. W. Powell, Geology of the Uinta Mountains, 1876. 346 FOREST PHYSIOGRAPHY The physiography of the Uinta Mountains is closely related to the geologic structure. The broad open valleys of gentle contour depend upon sculpturing of soft horizontal strata beneath a capping of harder beds. The steeply inclined strata on the flanks of the range have been eroded to the point where they stand out as a series of hogback ridges separated by parallel trough-like valleys. The streams run in a series of rapidly deepening canyons with nearly vertical walls 3000 to 4000 feet high, and are deepest where they cross the hogback ridges on the In the background. the Uinta fold ts supposed to have re. mained uneroded, while the foreground shows | the Uinta Mount- jains as they exist. al PY | Ti-ra-kav Plateau : zen) ea u NEEESLLEIEICE EEN ES OGMUITESSS ge ‘ea--Level SSK z = Fig 153 Fig. 111. — Stereogram and cross-section of the Uinta Mountains arch. (Dryer, adapted from Powell.) margins of the anticline. The axis of the Uinta uplift is near the northern margin and the canyons of the northern slope are therefore shorter than those of the southern slope. The inter-canyon spurs are broad and flat and are overlapped by gently sloping Tertiary beds in places up to elevations of 10,000 feet. The Green River enters the northern flank of the Uinta Mountains at the Flaming Gorge, then swings out again, only to return and cross the main axis of the range at the eastern end through the Ladore Gate or Canyon. Powell described the Green River as an excellent illustra- tion of an antecedent stream that had maintained its course across the Uinta Mountains during the period of their uplift.'. The later studies of Emmons show that. Powell’s hypothesis must be modified, for it involves physical impossibilities. 1 J. W. Powell, Geology of the Uinta Mountains, 1876. ROCKY MOUNTAINS. II 347 The upturned and truncated edges of the various formations of the Uinta arch are partly covered by overlapping Tertiary beds in a nearly horizontal position. These beds reach alti- tudes of gooo to 10,0co feet at various points on either flank of the higher western portion of the range. The Green River Canyon, at the Flaming Gorge, has an elevation in the eastern portion of the range of 7500 to gooo feet. From these conditions it seems clear that (1) the mountain-making movements which gave rise to the Uintas were accomplished principally at the close of the Cretaceous, though a small subsequent movement is allowable on strati- graphic grounds (each of the three series of Tertiary beds is marked by an erosion interval between successive beds and by a slight upturning on the flanks of the Uinta Mountains); (2) the Green River had its course directed as a consequent stream upon the surface of the overlying Tertiary deposits; (3) the dissection of the Tertiary beds allowed the river to be- come superposed upon the buried portion of the Uinta arch; (4) further erosion both within and on the borders of the range has intensified the topographic expression of the Uinta arch and to some extent obscured the origin of the course of the Green River.t FORMS DUE TO GLACIATION 2 In addition to the influence of alternating hard and soft beds on the valley outlines, glaciation has operated to widen and deepen the canyons and in many cases to give them characteristic U-shaped profiles. The floors of the basins in the western part of the range are about gooo feet in elevation and every large canyon that heads near this part of the crest has been glaciated. The glaciers in the central portion of the range were 20 to 27 miles long; elsewhere they were but 4 or 5 miles long. The ice slopes about Bald Mountain and Reed’s Peak near the western end of the range coalesced to form a great ice cap, Fig. 112. The greater portion of the divide was, however, not covered by ice and the loftier peaks rose above the snow-fields. The basins at the heads of the can- yons on the northern slope vary in area from 1 to 12 square miles, while many of those on the southern slope are 20 to 30 square miles in extent, a difference which is due to the fact that the southward-flowing streams have developed valleys in the plateau-like summit where the beds are nearly horizontal and where the conditions are therefore most favorable for glacial plucking and sapping and for the development of broad flat- bottomed cirques. The northward-flowing glaciers worked upon steeply inclined strata and were resisted by every hard stratum in the section instead of a single hard stratum; hence they were but little assisted by sapping, for the soft beds whose removal gave rise to the sapping of the harder beds above them soon dipped down beneath the plane of effective action. The result was not only a greater amount of work to be per- formed by the northward-flowing streams but also a restriction of the fields of nourishment which correspondingly decreased the intensity of glaciation. 1S. F. Emmons, Science, n. s., vol. 6, 1897, p. 131 et al. 2 W. W. Atwood, Glaciation of the Uinta and Wasatch Mountains, Prof. Paper U. S. Geol. Surv. No. 61, rgro. FOREST PHYSIOGRAPHY 346 (CAINS "JOO *S ‘A ‘poo JaIYy) ‘Sfolov[s AayAva Aq paiaAod wore quasoida1 pazop pur yor]q pros ‘sIdI9v[S s10}e]. Aq poxoAoo varv syuasasdar youjq POS ‘aus0}sSlo[q OY} Ul SUILJUNOJY LULA OY} JO SUIO}SAS JOIOe[H — ‘ZI ‘BIy colt 799} QOOT [eAs9} UT N07 MOD GPEGSTOL (tena nt td Cad nt be | BOLT JO [VS a VIN “HO LVSV AL 09 cd a) SULG Mh | ay ANIT aSVa ‘AS Deer tere ! DF 60L 1 02,601 ROCKY MOUNTAINS. II 349 FOREST GROWTH In the higher portions of the range the forest growth is extremely scanty, but in the lower valley basins there is a heavy growth of co- niferous forest. The upper limit of forest growth is about 11,000 feet, the lower limit on the southern slopes is beyond the base of the range in the Uinta Valley and is less than 7000 feet, while on the northern slopes it is gooo feet owing to differences of relief and to corresponding differences of exposure. The prevailing species are white pine, yellow pine, Engelmann spruce and Douglas spruce, besides species of lesser importance. “The view from one of the mountain lakes, with its deep-green water and fringe of meadow- land, set in the somber frame of pine forests, the whole enclosed by high walls of reddish- purple rock, whose horizontal bedding gives almost the appearance of a pile of cyclopean masonry, forms a picture of rare beauty.’ ! BIGHORN MOUNTAINS The Bighorn Mountains rise from 4000 to 5000 feet above the Great Plains to from 10,000 to 13,000 feet absolute altitudes. The range trends north-northwest in its northern portion, due north and south Fig. 113. — East-west section across highest part of Bighorn Mountains, showing extent of erosion and the contrast in dip on the two slopes of the range. in its central portion, and east and west where it joins the main Rocky Mountain System whose border ranges are known here as the Bridger Range and the Owl Creek Mountains.” It is in effect an extension or prodigious spur of the Rockies jutting far out beyond their border. The Bighorn Mountains are a great anticline lifted many thousands of feet and composed of a thick series of sedimentary rocks (Paleozoic and Mesozoic) and a central core of granite and schist (pre-Cambrian).’ The sedimentary rocks arch over the northern and southern portions of the uplift, and give rise to elevated plateaus about gooo feet high at the north and 8000 feet high at the south. On the divides these plateaus often present broad tabular surfaces though they are trenched by numerous large canyons. The central plateau terminates in high cliffs at certain places in the border region, but in others and especially 1S. F. Emmons, Geol. Expl. of the 40th Par. (King Surveys), vol. 2, 1877, pp. 194-195. 2 N. H. Darton, Geology of the Bighorn Mountains, Prof. Paper U. S. Geol. Surv. No. 51, 1906, pp. IO-II. 3 Darton and Salisbury, Cloud Peak-Fort McKinney Folio U. S. Geol. Surv. No. 142, 1906, Doss COlss 25535 FOREST PHYSIOGRAPHY 30° CAINS "OID *§ “f “‘UoJe™) “Yydessoj0yd oy} Jo saspia davys ay} ul UMOYsS A]1va]9 aav spaq pournzdn yj jo worssaidxo o1ydeisodo} ayf, “OAA\ ‘SUIeJUNOP[ WIOYSIg Jo oSpl1 JUWOIy 9UOJSOUT] JO Opis Jsva oy} UO vyRI}S Jo UINdn diveys — ‘bir “31,7 ROCKY MOUNTAINS. II 351 along the eastern side of the mountains it is flanked by a distinct ridge of limestone that rises slightly above an inner valley and slopes steeply toward the plains. These plateaus are covered by forests with many park-like openings and are extensively used as grazing grounds for cattle and sheep during the short summer season. The subordinate topographic features of the Bighorns depend upon local flexures on the flanks of the main uplift and a number of faults, the latter being especially marked northwest of Buffalo and on the eastern border of the range, where one of the faults has a throw of about gooo feet. The effect of the sharper flexing and of the pro- nounced faulting has been to increase the development of precipitous slopes, and to cause a greater development of hogback topography especially on the eastern side. The hogback ridges rise from t100 to 200 feet above the adjoining valleys, and are due to the outcrop of sand- stone of moderate hardness, while the valleys follow the outcrop of the “Red Beds” composed largely of soft shales. The general configuration of the central cluster of high mountains in the Bighorns is very rugged and presents some of the boldest alpine scenery in the United States. There are many precipices over 1000 feet high, particularly in the glacial cirques along the higher summits. Some of the cirques still hold glaciers; one on the eastern side of Cloud Peak is nearly one-half mile long. Extensive snowbanks remain the entire summer in many higher portions of the range.! GLACIAL FORMS At the time of maximum glaciation in the last glacial epoch the snow and ice accumulations of the Bighorns were two or three times as extensive as those of Switzerland to-day and the largest glacier was considerably larger than the largest existing Swiss glacier.” For thirty miles the crest of the range has been glaciated and signs of glaciation are everywhere abundant. The topographic and climatic conditions governing the distribution and growth of former glaciers were unusually favorable. Glaciers started from approximately one hundred sources, but in descending they combined in many cases, so that the number of separate glaciers or glacier systems was but nineteen. A specific instance is the Pear Rock Glacier, which started from not less than twenty sources. 1 Darton and Salisbury, Cloud Peak-Fort McKinney Folio U. S. Geol. Surv. No. 142, 1906, p. I. 2 Idem, p. 9, col. 4. Fig. 115. — Wall at head of cirque, upper end of a tributary of West Tensleep Creek, Bighorn Mountains. A. View from above rim, showing old rounded surface. B. View from below rim, showing granite walls nearly 1000 feet high. (Blackwelder, U. S. Geol. Surv.) 352 ROCKY MOUNTAINS. II 353 During both the advance and the retreat of the ice the number of separate glaciers was greater than during the time of maximum glacia- tion, for the ice was melted out of the lower glacial troughs, while it still lingered in the separate tributary valleys. The glaciers on the 107°007 SCALE OF MILES Zonet elope Ome ie Sie Oe 0 Fig. 116.— Glacier systems of the Bighorns at the time of their maximum development. (After Black- welder, U. S. Geol. Surv.) west side covered almost twice as great an area as those on the east and they were also somewhat longer, a difference due to the heavier precipi- tation on the west side (prevailing westerly winds) and the wider catch- ment basins. The surviving glaciers are now found wholly on the east side, a condition probably due to the better protection of the cirques 354 FOREST PHYSIOGRAPHY on that side from the sun and in part to the westerly winds which drift snow over the crest of the range and into the cirques and alcoves on the lower side. The lower limits of the glaciers were from 6500 feet to 10,500 feet, the necessary elevation for the generation of glaciers in the Bighorn Mountains during the last glacial epoch ranging from 9500 tO 115500: feet: Glacial cirques and hanging valleys are of frequent occurrence, though arétes and needles are absent because glacial sapping did not continue long enough to allow the meeting of glaciers on the divides.? The cirque walls have not been notably denuded in postglacial time be- cause of the compact and relatively structureless granite in which they were carved and in part because of the recency of formation. The cirques have been formed on a huge scale, for the original valleys were often far apart and neighboring cirques were thus developed without mutual interference on the lateral divides.’ Below the cirques are steep valleys whose gradients range from 600 to 800 feet per mile. Where the valleys cross the great central granite plateau they are shallower and more open; but where they cross the sedimentary rim they become narrow gorges. Some of the longest glaciers entered ‘these gorge-like marginal valleys, but none of them reached out upon the plains. Among the most important features of the present drainage are the large numbers of lakes due to glacial over- deepening or to the presence of morainic dams. Some of the terminal moraines of the Bighorns are of great size, the undulations of their surface reaching as high a value as 800 feet. The terminal moraine of North Fork is 300 feet above the level of the valley floor and its topography is very irregular.* FORESTS Nearly all the trees of the Bighorn National Forest (which includes most of the forested area of the mountains) are lodgepole pine (Pinus murrayana, locally called white pine), yellow pine, and jack pine. Another species is limber pine of scattered growth. Engelmann spruce occurs in some of the moister areas along the mountain slopes and the higher portions of some of the canyons. The dry or lower timber line 1 Darton and Salisbury, Cloud Peak-Fort McKinney Folio U. S. Geol. Surv. No. 142, 1906, p. 10, col. r. 2 F. E. Matthes, Glacial Sculpture of the Bighorn Mountains, Wyoming, 21st Ann. Rept. U.S. Geol. Surv., pt. 2, 1899-1900, p. 175. 3 Idem, p. 176. 4 Darton and Salisbury, Cloud Peak-Fort McKinney Folio U. S. Geol. Surv. No. 142, 1906, p. 11, col. 3. ROCKY MOUNTAINS. II 355 occurs at 6000 feet, and the cold timber line at 10,000 feet, so that the upper portions of the range do not generally support a timber covering of any importance, but on the whole expose bare rock-strewn summits. The wood of the white pine, the predominating species, is coarse-grained, knotty, and small. The lumber warps and cracks considerably, and it is therefore not regarded as valuable." 1 For detailed distribution of the timber of the Bighorns, etc., see roth Ann. Rept. U. S. Geol. Surv., pt. 5. CHAPTER XX ROCKY MOUNTAINS. III SOUTHERN ROCKIES IN contrast to the irregularly arranged mountains of the Central Rockies with variable trend, numerous offsets, and plain and plateau. interruptions are the northward-trending, somewhat regular, and con- tinuous mountain ranges that constitute the southern Rockies. There Scale of Miles D/ / /N Ha ifinte ty mp River Yampa ve iilaneas Danforth Oa Hills SS \ whit? White River ot Plateas Plateau ff S ) 12 Roan 9 S aos oS o. ?r 0 rs f i Wee ‘wer a ! \ MESA R lo \SOUTHY Fen nison : \ @ o “ncomp™ . San Miguel ,, Sage } Dolores {| Mts. aS ete Bear River Pllain Plat. | Mts. fe JLa Plata EI Late hie p = Park Plateau Mts. MESA VERDE Fig. 117. — Location map of the southern Rockies. 356 ROCKY MOUNTAINS. III 3547 appears to be a rather well-marked and consistently developed mountain plan in the eastern and central ranges of the district; the western- most ranges are of more diverse structure, topographic texture and origin, and occur in groups of. longer and shorter lengths arranged on the whole with less regularity. Fig. 117 represents the locations of the main topographic features of the region, which fall readily into five categories: (1) The eastern foothills, of varied origin and commonly exhibiting the hogback type of topography. (2) The Colorado or Front Range and the Wet Mountains, consisting in a broad way of a great anticlinal from whose central granite axis the flanking sedimentary beds dip east and west. (3) The Park, Sawatch (Saguache of some maps), and Sangre de Cristo ranges, all of which he nearly on the same meridian, trend north- ward, have roughly equivalent structures and topographic forms, and are parallel to the Colorado Range, from which they are separated by (4) a chain of ‘“Parks,’’ North, Middle, South, and Huerfano, four intermont basins of exceptional character, primarily of structural and secondarily of alluvial origin (San Luis Park, the largest of all the inter- mont parks of Colorado, lies between the Sangre de Cristo range and the southern continuation of the Sawatch Mountains— the Conejos, etc.). (5) Irregular mountain knots or groups of igneous origin which lie beyond the Park-Sawatch axes, such as the Elk, La Plata, San Juan, and Uncompahgre mountains. Inner crest of Outer Crest of red rock red rock Hcgback Fig. 118. — Generalized east-west section near Boulder, Colorado, showing the structure and topography of the South Boulder Peaks and the hogback ridge of Dakota Sandstone. (Fenneman, U. S. Geol. Sury, EASTERN FOOTHILLS HOGBACK TOPOGRAPHY The eastern foothills of the southern Rockies are generally formed upon a belt of sedimentary beds upturned at steep angles along the mountain slopes. The most characteristic feature is the hogback ridge, Fig. 118, developed in the form of long narrow ridges of harder 358 FOREST PHYSIOGRAPHY beds that stand like a fringing and often a serrated wall at a little distance from the base of the mountains. They occupy a belt about two miles wide involving portions of all valleys intervening between them and the main range. The easterly dip of the sedimentary beds of the foothills is quite general, and ordinarily varies from 35" toe3oy, though locally it is from 80° to go° Besides the general flexing of the strata, which is due in part at least to the uplift of the Colorado Range, there has been a considerable amount of minor crum- pling along lines parallel with the general trend of the range or diverging from it to some degree. The latter process re- sulted in a series of secondary folds reflected in the topography by 005 0/7 MIS ISLS successive offsets. In FOLDS EN ECHELON EA RE OF COLORADO RANGE, euch localiues hie ridges om §=3 mmg are arranged en échelon. rmo Hh » Longmont | ly sean l +# QUATERNARY LARAMIE FOX aoe CHAYDEN) ;MIOBRAR DAKOTA. JURASSIC RED BEDS ARCHEAN (HAYDEN) J Q8RARA (MORRISONYWYOMING) Their gener al appear- JURA-TRIAS Fig. 119. — Cross folds on eastern border of Colorado Range. ance is shown in Fig. IIQ, CU e Geel Sore) which represents a series occurring between St. Vrains and Cache la Poudre creeks, north of Denver. VEGETATION The vegetation of the Pueblo region is closely related to climate and soil, and in this respect and in its general character it is typical of the vegetation of the hogback ridges of Colorado. The uplands above 6500 feet bear a forest of yellow pine, and straggling pine growths occur on sheltered hillsides down to 5300 feet; in the same zone rocky slopes are sometimes covered with aspens. Hackberries and other hard- ROCKY- MOUNTAINS. III 350 wooded trees grow to moderate size in the moister valleys, while cot- tonwoods fringe the lower streams; juniper and pifion pine extend down to 5000 feet and are associated chiefly with dry rock-strewn ledges of resistant limestone and sandstone. OTHER TYPES OF FOOTHILL TOPOGRAPHY While the hogback type of mountain border or foothill topography is the predominating one there are a number of other types that require examination. They are not all confined to the southern Rockies, but are assembled here for comparison with the well-developed foothill types of Colorado. The hogback ridges are absent (1) where the harder beds rest directly upon the crystalline rocks of the mountain slopes, without the presence of a soft valley stratum between the crys- tallines and the hard sedimentary formations overlapping them, of which some of the foothills on the eastern border of the Bighorns are perhaps the best illustrations; or (2) where the upturned edges of other sedimentary (Mesozoic) beds are still covered by a capping of younger (Tertiary) beds in a horizontal position; or (3) where faulting has brought the sedimentary rock in a nearly horizontal position against the deformed crystallines as at Pueblo, Colorado. (4) The eastern border of the Rocky Mountains in Montana is developed upon strata that have been deformed in such a manner that they dip westward and present a reversed border, (a) a single, steep, precipitous, eastern scarp overlooking the Great Plains, such as marks the eastern front of the Lewis Mountains, Fig. 92, or (b) a series of scarps and flats formed upon alternate soft and hard layers— short, steep, eastern slopes and long, gentle, mountainward or western slopes, Fig. 93. In the Bighorn Mountains and in the Colorado Range, on the other hand, the marginal strata in general dip steeply eastward, with the consequence that typi- cal hogback ridges are characteristic. (5) Lava flows have occurred locally and have caused the formation of a mesa type of topography or (6) volcanic cones surmount the eastern border. Besides the hogback type six types of foothill topography are thus distinguishable along the eastern border of the Rockies. Instead of being everywhere formed simply upon the basset edges of the sedimentary beds exposed along the mountain flanks, the border has been formed in diverse ways and exhibits diverse topographic features. While each of the types indicated has distinctive features the first five types are so closely related as not to require detailed explanation. The sixth type— the igneous border —is well developed in the southern Rockies and has rather complex features that merit further consideration. 360 FOREST PHYSIOGRAPHY IGNEOUS BORDER SPANISH PEAKS In southern Colorado and locally elsewhere in that state, as at Golden, the strata have been deformed or to some extent concealed and the border given an unusual character by the occurrence of igne- ous masses which in the case of the Spanish Peaks consist of volcanoes and associated dikes and lava flows. The type of mountain border exhibited is wholly unlike that found anywhere else along the eastern face of the Rockies from Alaska to New Mexico. The most prominent feature of this kind is the volcanic nucleus known as the Spanish Peaks. The two culminating points, West Peak Fig. 120.— West Spanish Peak, from the northwest. One of the large dikes of the region is seen in the foreground. (U.S. Geol. Surv.) and East Peak, are about 2 miles apart and have elevations of 13,623 feet and 12,708 feet respectively. They rest upon a small platform that descends eastward gradually and terminates in an irregular and deeply dissected line of steep bluffs that descend abruptly 500 feet to the gently rolling plains. The average elevation of this platform is about 7500 feet. Its surface is a succession of mesa-like ridges and narrow valleys of extremely rugged development. The peaks themselves consist of stock-like masses occupying nearly vertical fissures, and from them the flanking strata dip steeply away with various structural complexities. The general ruggedness of the country about them is emphasized by ROCKY MOUNTAINS. III 361 many dikes which have a rude radial arrangement with respect to the peaks which were the centers of eruption. They were formed in crustal fractures related to the deformation of the surface strata, deformations produced by the intrusion of great masses of igneous rock.! They are from 2 to too feet in thickness. Some of them are practically vertical and project above the surface as smooth wall- like masses from 50 to too feet high. Their crests are sometimes straight for long distances, although curved crests are most common. They range in length from a few hundred yards to 1o and 15 miles, and intersections are common.’ The upland portion of the region adjacent to the Spanish Peaks is rather heavily timbered and the eastern borders of the plateau support a dense growth of pifon and juniper, scattered pines, and scrub oak, with occasional park-like openings. The more elevated central, south- ern, and western borders support forests of pine; a certain amount of spruce, fir, and aspen grows at the base and on the lower slopes. The summits of the peaks are from 1000 to 1500 feet above the cold timber line. The plains eastward of the plateau upon which the Span- ish Peaks stand are practically destitute of timber except for fringes of cottonwood along the running streams. The whole plateau portion is subject to frequent summer showers. The summits of the Spanish Peaks are rarely free from snow for more than two months in the year, while along the narrow valleys cut below the level of the plateau irri- gation is necessary, as also upon the surface of the plains country toward the east.® NORTH AND SOUTH TABLE MOUNTAIN Among the various types of foothill topography one of the most important is the flat-topped mesa type, which is well shown in the vicinity of Golden. Here the hogbacks completely disappear and in their place are two basalt-capped, sedimentary masses known as North and South Table Mountain. They were originally continuous but have since been cut through and isolated (Clear Creek). The basalt was extruded after the strata had been flexed into steep attitudes and some- what eroded. It consists of two fissure flows with no sedimentary rock between them.* They protect the underlying beds from erosion; their present elevation indicates the degree of dissection that has taken place in the vicinity since their formation. 1 R. C. Hills, Spanish Peaks Folio U.S. Geol. Surv. No. 71, 1901, p. 3, col. 1. 2 Idem, p. 3, col. 4. 3 Idem, p. 1, col. 1. 4 Emmons, Cross, and Eldridge, Geology of the Denver Basin in Colorado, Mon. U.S. Geol. Surv. No. 27, 1896, p. 285. 362 FOREST PHYSIOGRAPHY FOOTHILL TERRACES The foothill terraces are among the important features on the eastern border of the Colorado Range. They are mesalike remnants of stream terraces carved in rock, Fig. 118, standing out with special prominence along the base of the foothills south of Boulder where they rise from too to 300 feet above the lower plains. They vary in width from a fraction of a mile to three miles, and descend eastward with slopes varying from 1° to 10°. They appear to be due to a leveling of the rock margin of the plains by meandering streams and the deposition upon the plains surface of a thin sheet of unassorted gravel.! The terraces always slope toward the streams, with which they once had flood-plain relations. COLORADO OR FRONT RANGE The Colorado or Front Range is the easternmost of the imposing chains that form the southern Rockies. Its elevation above the surround- ing country is represented in Fig. 121. On the north the Colorado Range is a high, heavily timbered, plateau-like range separated from the Medicine Bow Mountains by the Big Laramie River, and from the Laramie Mountains by the North Fork of the Cache la Poudre; but throughout most of its extent it is alpine in character, with sharp serrated summits bordered by great glacial amphitheaters.2 The broad anticlinal of the range has a very flat summit arch, with increasing dip toward the flanks of the anticline on the borders of the range.*? The rocks of the Colorado Range are almost exclusively crystalline granites, gneisses, and schists.4. The topo- graphic features are not uniform throughout and we shall therefore describe them by groups. West and northwest of Denver and north of Arapahoe Peak, the serrated crest of the Colorado Range is exceedingly formidable, with many peaks rising to 13,000 and 14,000 feet. Even the passes are high and difficult, and on either side are steep-sided and deep mountain gorges. The irregularity of the mountains and their steep descents are suggested by the northeast face of Long’s Peak (14,270), which consists of an almost perpendicular cliff over 3000 feet high which 1 N. M. Fenneman, Geology of the Boulder District, Colorado, Bull. U. S. Geol. Surv. No. 265, pp. 13-15. 2 S. B. Ladd, U. S. Geol. and Geog. Surv. of Colorado and Adj. Terr. (Hayden Surveys), 1873, p. 436. 3 Clarence King, U. S. Geol. Expl. of the 4oth Par. (King Surveys), vol. 1, 1878, p. 21. 4 Arnold Hague, Rocky Mountains, U.S. Geol. Expl. of the goth Par. (King Surveys), vol. 2, TOD 22). 363 Ill ROCKY MOUNTAINS. (‘sAaaAIng 3uly) ‘aoutaoid urejunoyy ANIOY oY} Ul sasuvi JUIULwWoId aAy jo sajyord jeurpnysuoy — izr “Bry *3[9A9]| Surpunosans aaoqe saduva Jo UOIVAD|Y YU} I9}YSY *]OAo] Vag - sauty ose ‘L334 008 SLNASSYdSY AOVdS HOWS —3IVOS 1VOILESA ¢ GEOb JET WO fsa Ay 07 Ise MOAT LL Viel SONVA VININ Y/ YW) Zl mm pVyjjyyyyys ee “A BEE é LLP “g) om usps ie —Watrah BE yoo a uv 4 PO Cre aati i 2 at eno eeu fe cars pony 3 ee Wet 89,801 “Buoy SECO Usa 0) ur LY ot ad °° 75 “Ad * ‘SIW MOS JANIDIGSW YL, /S$S,000 “S007 YY OF 90L * Wy YY ify) LL LLL, /. ly Z Kile YLLLL Me eS A } oN gw a ye “ WAT god' oa s0 aw uae SPE OST ov (EN yo n KLM LL, (9883) qn10 —___1d wilh _ a0 ot wi 8 9,00-0°NY (100 ¥ Sud Cra FOREST PHYSIOGRAPHY ; ule ANI) Te gas If pif} aa GEOLOGIC SSO Mi ON ~ .7 op STRUCTURESECTION = (3 = NY } OF THE Bee Se N pssiata BES Sh Mi DENVER BASIN 2uk cae CF COLORADO SIREN Efe =i ay . TTT ‘ Carrs Se Xe ———} | ‘ NOOO, aS Scale of Miles M00 857) Bl rT dl -~ ~*) =). =H hN p Nw 0 4 1 2 S925 C/N: ©} 6 077 (0 S23 185 {/ My NCA ueTe ; a, = ey | IN Wi +9) cil an | Sar aa Var ae Coearennne oe oe f RATAN SN as Se, — i. me Ses | 2 3 Tey 7 SNS IS, OM Contour Interval 250 f MANS = |I] WYOMING Yj FORMATION FORMATION FORMATION IN FORMATION Ewe] FORMATION WY): BASALT (UPPER DIVISION ) (Lower Division) Hise PRE-CAMBRIAN GRANITES GNEISSES ~—’! AND METAMORPHIC ROCKS ERRAUETS 2 Fig. 122. — Section on the common border of Great Plains and Southern Rockies, showing structure and topography and the lava-capped mesas near Golden, North Table, and South Table mountains. (U.S. Geol. Surv.) extends from timber line (11,000 feet) to the mountain summit. Its eastern face presents an almost continuous line of steep-walled amphi- theaters; the western side of the range is a zone of high mountains 5 to 10 miles across which have steep but not precipitous slopes that descend to the valley of the Grand and are cut by profound canyons. ROCKY MOUNTAINS. III 305 Southward for 12 miles from Arapahoe Peak the topography changes. The crest presents a very uniform ridge but little above timber line. Like the section north of Arapahoe Peak the eastern face of this sec- tion is diversified by amphitheatral valley heads between which are broad rounded spurs of such regular descent that wagon roads ascend some of them. The western face of this portion of the range is marked by massive spurs with rounded forms not separated by canyons.! The whole eastern border of the Colorado Range is usually well defined and even sharply defined, and all along it the mountain spurs end abruptly. They rise from 5000 or 6000 feet, the level of the border of the plains, to 8000 feet at the foot of the main range, and involve a belt about 8 miles wide. The streams draining the border have cut deep parallel valleys and the intervening ridges therefore have an east- west alignment. They are not sharp but massive and have rounded or level summits and steep margins. Their crests fall into a general level in a quite remarkable manner. On the western border of the Colorado Range the descent is also rather abrupt to a general plateau between 8500 and 10,000 feet high which stretches westward for many miles.” These border plateaus have been studied in detail in the Georgetown region west of Denver and their topographic relations clearly defined. It has been found that the Colorado Range here exhibits three sets of topo- graphic forms: (1) an old, mountainous upland, (2) young V-shaped valleys incised in the upland, (3) glacial cirques developed in the valley heads. The old upland surface, but little modified by recent erosion, occurs in remnants preserved on the ridge crests, which in part accounts for their regularity of elevation, form, and alignment. This mountainous upland was an ancient land surface whose relief in the preceding topo- graphic cycle would be adequately shown by filling the deep valleys cut into it.2 Dome-shaped mountains and broad soft-contoured ridges stood where sharp peaks and more rugged ridges now are, and the val- leys between were broader and less steep than those of the present streams. The surface was well adjusted to that of the underlying rocks and to their varying resistances to erosion. Precipitous slopes were rare and occurred under special conditions, as on the southeastern border of the Alps Mountains of Colorado. Shallow valley heads led 1 F. V. Hayden, U. S. Geol. and Geog. Surv. of Col. and Adj. Terr., 1873, pp. 86-87. 2 Idem, pp. 88-80. 3 Spurr and Garey, Geology of the Georgetown Quadrangle, Prof. Paper U. S. Geol. Surv. No 63, 1908, pp. 31-36. N.M. Fenneman some time ago called attention to the fact that these features appear to indicate an imperfectly base-leveled surface that has been trenched by streams invigorated by recent uplift: Geology of the Boulder District, Col., Bull. U. S. Geol. Surv. No. 265, 1905. 306 FOREST PHYSIOGRAPHY down gently from the dome-shaped mountains of the old upland and joined wide gentle valleys that were broad and basin-like. All these members of the ancient stream systems had normal profiles and no reversed slopes existed; consequently lakes did not then exist. -Although both normal and glacial erosion have modified this old upland, rem- | —Ss LH ox a O eS ——— fi eS — 0009 > Se /-11000 Sp 3 iG 2009 if PRESENT TOPOGRAPHY, THE OLD MOUNTAINOUS UPLAND RESTORED Toapalapenlin alee! Lee 1 0 1 2 3 4 5 miles 1 0 1 2 3 4 5 miles Contour interval 500 feet t 1906 Contour interval 500 feet Tig. 123. —Old mountainous upland of the Georgetown district, Colorado, and present topography. (Spurr and Garey, U. S. Geol. Surv.) nants of it may still be seen and such remnants bear residual elevations and in many places are covered by deep soil, conditions that were probably general over the larger portion of the area before uplift and I glaciation took place, so that rock exposures were comparatively un- i common except on the steepest slopes.! (i 1 Spurr and Garey, Geology of the Georgetown Quadrangle, Prof. Paper U. S. Geol. Surv. No. 63, 1908, p. 32. ROCKY MOUNTAINS. III 367 The mountainous upland of the Colorado Range, formed in late Tertiary time, was then uplifted and tilted eastward so that valley carv- ing was begun. Canyons of considerable size were opened up before the glacial period and these the streams are still enlarging. Clear Creek is in places 23 miles wide, while the gorges along the Deer, Elk, and Bear creeks, all in the eastern slope of the range, are in general not more than one-half mile wide. The valley walls approach each other upstream and the valleys finally become narrow V-shaped notches in the old upland surface. Rock projects from the valley sides and huge bowlders lie along the courses of the swift streams. The old upland surface is comparable with similar uplands (a) in the vicinity of Pikes Peak,! (6) in the Sierra Madre Mountains of south-central Wyoming,? (c) remnants of an ancient surface widely distributed over other por- tions of Colorado,* and (d) mature profiles widely developed in other portions of the Rockies as in western Montana, p. 318. The character of the ‘old mountainous upland is brought out clearly in the maps of the Hayden Surveys of 1877, where the summits of the long table- topped spurs of various plateau and mountain groups appear to indi- cate a rather general development of moderate slopes. All of these surfaces are approximately contemporaneous and of late Tertiary age. The mountainous surface is now exposed only in remnants on the spur tops and in certain of the unglaciated valleys. The present-day expression of the detailed topography of the Colo- rado Range west of Denver is affected rather intimately by the char- acter of the rock wherever dissection has partly destroyed the older graded surface. The country underlain by schists has very distinctive topographic qualities. Where the dip of the schistosity is low, smooth- contoured slopes are developed upon the exposures; where it is approxi- mately vertical the topographic forms are craggy and serrate as on the north side of Chief Mountain. The gneisses weather into exposures which at a distance have the bedded aspect of sedimentary rocks except where there is a prominent development of joints. The rugged crags found almost continuously along stream valleys carved in gneiss are controlled in their minor details by rectangular joint systems and have very characteristic features. Turrets and buttresses and castellated forms are seen throughout the area covered by the granite-pegmatite, which is one of the most resistant massive rocks of the region. - 1 W..Cross, Mon. U.S. Geol. Surv., vol. 27, p. 202. 2 A. C. Spenser, Prof. Paper U. S. Geol. Surv. No. 26, 1904, p. 12. 3 F. V. Hayden, U. S. Geol. and Geog. Surv. of the Territories, Atlas of Colorado, 1877, Sheets 5, 6, 7. - 3608 FOREST PHYSIOGRAPHY Below timber line the granite weathers into typical dome-shaped forms, the domes being from 200 to 300 feet high and from 300 to 400 feet in diameter; above timber line the domes are absent owing to the strong temperature changes and their shattering effect.1 - In the Pikes Peak district of the Colorado Range the chief top- ographic feature is a great plateau developed upon granite and gneiss and modified by large extrusions of volcanic rock that form mountains upon it. Sedimentary beds, formerly of greater extent, are now con- fined chiefly to the borders of the tract and to small local basins on the plateau.” GLACIAL FEATURES The higher parts of the main ranges of Colorado were for the most part covered with ice during the last glacial epoch, only the narrow crests of the inter-valley ridges projecting above the glacial cover. The ice was deepest and most powerful in the larger mountain val- leys down which it extended as valley glaciers for great distances, in some cases out over the piedmont plains or plateaus below.? The number of glaciers which existed in the Colorado Range during the glacial epoch may be appreciated by the fact that within the Lead- ville quadrangle alone, p. 371, there were 26 large glacial systems and 11 smaller ones, or 37 in all, in an area of 945 square miles.* The glaciers ranged in elevation from a minimum height of 8800 feet to a maximum height of 13,700 feet. Since-many peaks of the Park Range reach elevations approximating 13,000 feet it follows that all the larger valleys were occupied by glaciers. The larger and more vigor- ous glaciers developed on the eastern slopes, a condition due probably to the greater size of the preglacial valleys rather than to any advan- tages of exposure or precipitation. As a result of glaciation in the Colorado Range cirques and glacial lake basins were formed. The cross sections of the valleys were developed into a pronounced U shape, so that a shoulder now occurs between a lower, younger, and slightly modified glaciated portion and an upper, older, unglaciated portion. The floors of both cirques and glaciated valleys are very uneven and 1 Spurr and Garey, Geology of the Georgetown Quadrangle, Col., Prof. Paper U.S. Geol. Surv. No. 63, 1908, pp. 35-36. 2 W. Cross, Pikes Peak Folio U. S. Geol. Surv. No. 7, 1894, p. 3. 3 The extent of any glacier depends upon the following features: (a) the height of the sur- rounding mountains; (b) the shape and size of the catchment basins; (c) the number and size of the tributary valleys; (d) the slope and shape of the valley floor; (e) the exposure; and (f)' the precipitation. 4S. R. Capps, Jr., Pleistocene Geology of the Leadville Quadrangle, Col., Bull. U. S. Geol. Surv. No. 386, 1909, p. 9. ROCKY MOUNTAINS. II 360 smooth rock ledges protrude above the till. Lateral moraines, some- times double, form prominent narrow ridges of till on either side of the valley walls and are here the most striking types of glacial deposits. The terminal moraines have an irregular surface marked by hillocks, ridges, and depressions and many of them have very steep fronts over which the streams descend in a succession of waterfalls. The cirques are the most striking feature of glacial erosion in the region; the deepest are in the vicinity of Mount Evans, where one cirque wall rises 1700 feet in less than a mile. The steeper cirques have bare, rocky, unscalable walls and frequently have postglacial talus accumulations along their bases. Some of them are compound, as in the case of the three well- defined cirques at different elevations in the valley of the stream that heads in Summit and Chicago lakes, a mile west of Mount Evans. In a number of cases a knife-edge ridge or aréte was formed by the headward erosion of two cirques on opposite sides of a divide in the mountainous upland. TREE GROWTH The main portion of the Colorado Range was heavily timbered below the cold timber line (11,250 to 12,000 feet) when the valleys of the S ng Eclotedo awatch 9 ange Wasatch Mts. < Mts) ee is 5 23 Be ete és ? 1020002. Great Plains 2 7 64 NSF = 0 miles Sea Level 510 Fig. 124. — Topographic profile and distribution of precipitation across the Wasatch and the southern Rockies. region were first settled. At altitudes over 8500 feet western yellow pine and Douglas spruce grow on dry areas. On higher and moister ground lodgepole pine and Engelmann spruce are found; these are superseded in turn by dwarf spruce and finally near timber line by mountain white pine. Large areas have been burned over and are now covered with second-growth lodgepole pine and aspen. Above timber line there is an alpine flora of many species.! PARK RANGE North of the Grand River Valley the Park Range is not high and rugged but a great, rolling, heavily timbered, even-contoured plateau of 1 Spurr and Garey, Economic Geology of the Georgetown Quadrangle, Col., Prof. Paper U.S. Geol. Surv. No. 63, 1908, pp 30-31. 370 FOREST PHYSIOGRAPHY crystalline rock with massive marginal slopes. For 50 miles south of the Grand River Valley its character is entirely different. The sum- mit ridges are rugged and sharp and bordered by great amphitheaters with steep, cliffed sides. Deep glacier-scored canyons alternate with the secondary ridges; at their mouths are belts of morainic drift formed by ancient glaciers whose courses were the present valleys or canyons and whose névé fields occupied the existing amphitheaters of the valley heads.! Glacial action was carried so far that parts of the range were thoroughly skeletonized. Some of the eastern spurs are flat-faced and have gentle descents. At the higher elevations the western slopes are precipitous being formed upon hard crystalline rock, but at lower ele- vations where sedimentary rock occurs they become gentler. On both flanks of the range the sedimentary formations seldom rise more than a few hundred feet above the bordering plains and plateaus. The cul- minating summits of the range are of granite.2 In structure the Park Range is a great overthrust extensively eroded. The movement was directed westward, hence the western border is abrupt, for it has been developed upon the edges of the overthrust beds in a manner in some respects similar to the eastern border of the Lewis range of Montana (p. 311). The highest peaks are much lower than those of the Colorado Range; Mount Zirkel alone reaches above 12,000 feet, and this in spite of the equivalent elevations of the main summits of the two ranges.* TREE GROWTH The Park Range is covered with large timber similar to that found on the Colorado Range. The main types are pines above and aspens and small low trees along the lower or dry timber line. In the local alluvium-floored basins timber is wanting, as in Egeria Park and the parks along the Yampah, except for a fringe of cottonwoods bordering the streams.° SAWATCH RANGE The Sawatch Range extends for over 80 miles from the Mountain of the Holy Cross (lat. 39° 28’) southward to the San Luis Valley. “For this entire distance [it] literally bristles with lofty points about to of which rise above 14,000 feet and many more are 13,000 feet above sea 1 Hayden Reports, 1873, pp. 178, 188, 180, 436, 437- 2 Idem, pp. 65-66. 3 Tdem, 1874, D. 71. 4 Arnold Hague, U.S. Geol. Expl. of the 4oth Par. (King Surveys), vol. 2, 1877, pp. 130-132. 5S. B. Ladd, U. S. Geol. and Geog. Surv. of Col. and Adj. Terr. (Hayden Surveys), 1874, Pp. 441. ROCKY~ MOUNTAINS. III 371 Pearl? Cr 10008 54900 ne. #000, C ES FER f #440007 7g 0 ®P ed, [05 nh oe Qa RY? 7 == ale Se Giemsa ang ms Senin Seale of Miles 0 1 2 3 4 5 6 Fig. 125. — Extent of the former glacier systems in parts of the Park (east) and the Sawatch (west) ranges of Colorado, Leadville quadrangle, (After Capps, U. S. Geol. Surv.) 372 FOREST PHYSIOGRAPHY level.” It has a rather symmetrical outline, and its pointed summits vary but little in either form or height.t On the east is the valley of the upper Arkansas; on the west is the valley of the Gunnison. Like the neighboring ranges on the north and east it is a great anticlinal uplift whose central axis is composed. of crystalline rocks from which sedi- mentary formations dip away at various angles. This type of structure is, however, complicated on the southwestern flank of the range by volcanic flows which wrap about the Sawatch crystallines, as the Cachetopa Hills— a portion of the great igneous field of the San Juan region. Because of this and other structural irregularities, the range is not so simple as the Park and Colorado ranges; and is still more irreg- ular in the southeastern extension of this mountain axis known as the Sangre de Cristo Mountains. The higher portions of the Sawatch Range have been glaciated in common with the other high ranges of Colorado. Glacial cirques or headwater amphitheaters were developed and peaks trimmed to sharp, sometimes pyramidal outlines. La Plata peak in the Sawatch Range has been glaciated to such an extent that it now stands as a sharp peak between the encroaching heads of three glacial channels. Its steep slopes merge into the steep slopes of the glacial cirques that occur northwest, southwest, and east of the peak. The forms of the mountain are all sharp in contrast to the rounded forms of the unglaciated lower portions of the adjacent mountains. Mount Elbert may be cited as a type of peak in which the glacial cirques have not encroached so far and where the summit of the mountain still retains a massive dome-like outline such as La Plata Mountain had before the erosion on its flanks had been carried so far.2 SANGRE DE CRISTO RANGE The Sangre de Cristo Mountains, Fig. 117, rise abruptly on the east from Wet Mountain Valley and Huerfano Park, among the larger parks of Colorado, and on the west from the San Luis Valley. They are from 12,000 to 14,000 feet high with from 3000 to 5000 feet relative altitude. The range follows an almost straight line for 4o miles and presents a very imposing sight, the dark color of the rock bringing out its relief to advantage. Its average width is not much over to or 12 miles, which is small compared with its length and altitude. The range extends southward to Sangre de Cristo Pass, and beyond this point the axis of elevation extends to and beyond the New Mexico boundary to Santa Fe, where it ends in a number of minor ranges. 1 Hayden Surveys, 1874, p. 54. 2 W. M. Davis, Glacial Erosion in the Sawatch Range, Col., Appalachia, vol. io, 1890, Pp. 403. ROCKY MOUNTAINS. III OS North of Poncha Pass the Sangre de Cristo Range is a true sierra. The Sierra Blanca group of peaks, for example, have a serrate and jagged crest line; their precipitous front rises abruptly from the flat San Luis plain to heights of more than a mile, thus giving the range a bold and majestic appearance.! The Sangre de Cristo Range is a great and somewhat complex anticline. In general the axis of the range consists of intrusive granite flanked on both sides by conglomerate chiefly and also by sandstone, shale, and limestone. Faulting has locally caused the sedimentary formations to abut squarely against the basal granite. In places the faults trend at an angle with the main axis of the anticlinal uplift, a structure that causes the topographic condition of offsets that diversify the mountain border.? Local variety is given the border relief by dikes of igneous rock which traverse the flanking sedimentaries. The streams have cut deep gorges in the upturned sandstones and limestones; steep bluffs occur along the valley sides and only an occasional patch of flat land occurs on the valley floors.* WESTERN BORDER RANGES The eastern and central divisions of the southern Rockies are char- acterized by the presence of granitic masses and extensive areas of meta- morphic schists typically represented in the Colorado and Park ranges. But in the westernmost ranges of the district vast quantities of volcanic material have been extruded, which have in many notable instances pro- foundly modified the topography. The San Juan, La Plata, Needle, and Elk mountains may be taken as typical illustrations. They are so rugged and access to them is so difficult that permanent settlements have been formed in very few places. There are few trails, and at great distances from each other are scattered prospectors’ cabins, now commonly deserted.’ 1 C. E. Siebenthal, Geology and Water Resources of the San Luis Valley, Col., Water- Supply Paper U. S. Geol. Surv. No. 240, 1910, pp. 34-35. 2 Idem. 2 For structural features see J. J. Stevenson, The Geology of a Portion of Colorado, U. S. Geog. Surveys West of the rooth Meridian (Wheeler Surveys), vol. 3, 1875, pp. 490 et seq. 4 F. M. Endlich, U.S. Geol. and Geog. Surv. of Col., and Adj. Terr. (Hayden Surveys), 1873, pp. 323 ff. 5 So little known is the region that during the survey of the Needle Mountains quadrangle, in 1900, a large number of peaks and other prominent features were given names for the first time. : 374 FOREST PHYSIOGRAPHY SAN JUAN MOUNTAINS The term “San Juan region,” Fig. 117, is generally applied to a large tract of mountainous country in southwestern Colorado and an ad- jacent zone of undefined lower country bordering it on the west and south. The greater part of the district is a deeply scored or dissected plateau drained on the north by branches of the Gunnison, on the west by branches of the Dolores and San Miguel, on the south by the San Juan, and on the east by the Rio Grande; the eastern part consists of high tablelands that represent old plateau surfaces; the San Juan Mountains proper extend from the great plateau region of Colorado and Utah on the west to San Luis Park on the east and from the canyon of the Gunnison on the north to the rolling plateaus of New Mexico on the south. They have a length of 80 miles east and west and a north- south width of nearly 4o miles, their summits forming a great group rather than a range. The elevations of the highest peaks exceed 14,000 feet, and hundreds of peaks exceed 13,000 feet. Some of the valleys in the heart of the mountains haye been cut down gooo feet or more, and the configuration, especially on the west, is extremely rugged and presents an almost infinite variety of slopes. Several of the small groups of high peaks in the border zone have special names, such as Needle Mountains on the south, La Plata and Rico mountains on the south- west, and San Miguel Mountains on the west. The geologic structure and to a certain extent the topographic character of the bordering groups are sufficiently unlike the San Juan Mountains to constitute them subordinate topographic divisions. The southwestern border of the San Juan Mountains is marked by bluffs with vertical faces where the great upland crowned by moun- tains descends to the Navajo, San Juan, Piedra, and adjacent stream courses, Fig. 117. When viewed from this direction the upland appears as a rugged range, but at the general summit level its true character as a great plateau crowned by isolated summits becomes apparent. The relatively unbroken character of the surface over a considerable number of summit areas prevents thorough drainage, and swamps abound above timber line. A special feature of the San Juan region remarkably developed in tke San Juan Mountains and in many places in the Needle Mountains is the large number of landslides determined by the alternation of soft shale and beds of limestone or sandstone. They consist of great blocks of rock tilted at various angles, with fine material collected in the intervening spaces, and have very uneven surfaces without regular ROCKY- MOUNTAINS. III 375 drainage systems. Stagnant pools and morasses alternate with deep trenches. The slides occur at the heads of the ravines and on the sides of the steep ridges separating the smaller valleys." The landslides of the Telluride and Rico districts of the San Juan region are supposed to have been caused in part by earthquake shocks on such a prodigious scale that the sliding and fracturing of solid rocks occurred. Possibly also the melting of supporting ice masses after they had deepened the valleys and steepened the valley slopes may have Fig. 126.— Landslide surface below Red Mountain, near Silverton, Colorado. added to the effect of the structural conditions.” Earthquake shocks have occurred in recent years in the Telluride and Red Mountain dis- tricts within the landslide area, and fresh fractures have also been observed in the Red Mountain mines.* In the Telluride region the rocks involved in the landslides have been those of the volcanic series which rest upon soft Cretaceous shales.* 1 Ernest Howe, Landslides in the San Juan Mountains, Colorado, Prof. Paper U.S. Geol. Surv. No. 67, 1909. 2 Cross and Howe, Silverton Folio, Col. U. S. Geol. Surv. No. 120, 1905, p. 25, col. 1. 3 H. C. Lay, Trans. Amer. Inst. Mining Engineers, vol. 31, 1902, pp. 558-567. 4 F. L. Ransome, A Report on the Economic Geology of the Silverton Quadrangle, Col., Bull. U. S. Geol. Surv. No. 182, 1901, pp. 27-28, et al. 376 FOREST PHYSIOGRAPHY RAINFALL AND VEGETATION The San Juan Mountains are surrounded by arid country, but they themselves have an abundant precipitation, the higher peaks and basins being seldom entirely free from snow. The abundant water supply supports a heavy forest growth in many localities upon the western and northern slopes. The upper timber line lies between 11,500 and 12,000 feet, and as a consequence large areas in the lofty interior of the San Juan Mountains are without tree growth, although a low alpine flora is found in favored situations. Spruce and aspen cover the higher slopes, and below them white pine, scrub oak, phion pine, and cedar are found. On the great volcanic plateau of the San Juan region the precipitation is less and supports only an herbaceous vegetation. Prominent yet low cone-shaped peaks present a more desolate appear- ance and are sometimes so strewn with rock fragments as not to support plants of any sort. High plateaus of volcanic rock appear in places with a grassy vegetation so different from the forests on the older sedimentary and igneous rock to the east and locally in the deep canyons of the vol- canic area, as to form in a certain sense a guide to the geologist.' LA PLATA MOUNTAINS The La Plata Mountains are located between the well-watered, well- forested San Juan Mountains and the arid mesa, plateau, and canyon country that extends into Utah, Arizona, and New Mexico. The La Plata Mountains are a dissected dome and their drainage systems have a radial arrangement. Many of the peaks rise a thousand feet or more above timber line and are characteristically rugged” On the west the slopes descend rather steeply 4000 or 5000 feet to the rolling plain known as the Dolores Plateau, deeply dissected by a number of deep canyons, Fig. 78. The simple character of the La Plata Mountains is complicated on the western and northwestern sides by intruded igneous rocks which occur in soft shales overlying sandstone, the intrusive rock having prevented the uniform erosion of the shales. On the southwest the dip of the slope of the sandstone is uninterrupted between 8000 and 10,000 feet, and the descent in this quarter is regular and broad. The intrusive rock is for the most part porphyry, and the masses are large enough to cause peaks or ridges which appear as irregularities in the 1 F, M. Endlich, U. S. Geol. and Geog. Surv. of Colorado (Hayden Surveys), 1873, p. 306. 2 Cross and Spencer, La Plata Folio U. S. Geol. Surv. No. 60, 1899, p. 1. Fig. 127. — Looking down La Plata Valley from the divide at the head of the valley. The steep slopes \ are characteristic. Landslide in left foreground. (La Plata Folio U. S. Geol. Surv.) Fig. 128. — Western summits of La Plata Mountains from the divide at head of La Plata Valley. This view is panoramic with the view above. It shows the rugged summits within the eroded volcanic stocks of the mountains. Diorite Peak on left. (La Plata Folio U. S. Geol. Surv.) ; 377 ao 378 FOREST PHYSIOGRAPHY general slope of the dome. The porphyries occur as sheets or small laccoliths;! later intrusions took place which cut the sedimentary beds and the sheets of porphyry forming stocks which constitute several of the highest peaks of the La Plata Mountains. : In connection with the formation of the dome of the La Plata Mountains a number of faults were developed but these are neither numerous nor important. The upthrust is commonly on the side near the center of the dome and has resulted in the formation of a number of steep slopes facing southward? The arid plains or plateaus rising west of the La Plata Mountains support a growth of white pine, pinon, and cedar, transitional between the barren expanses of the plateau and the forested slopes of the La Plata Mountains. In general the heaviest timber grows on the western and southern slopes since it is from those directions that the winds commonly blow. The forests extend up to 11,500 and 12,000 feet; the growth consists of spruce, fir, and aspen; above the timber line there is a scanty vegetation of alpine character.® NEEDLE MOUNTAINS The rocks composing the Needle Mountains are chiefly granite, schist, and some quartzite. Four of the summits exceed 14,000 feet in eleva- tion and many exceed 13,000. Some of the mountains are deeply dis- sected and extremely bold. North of the Needle Mountains is the great curve of the Grenadier Range, which is chiefly of quartzite; on the north and east the surface is almost entirely covered by volcanic or other rocks; on the south sedimentary rocks were laid down upon a pre-Paleozoic land surface of moderate relief now exposed (by the removal of the sedimentary cover) in the form of a southward-sloping tableland consisting of isolated mesas between deep-cut valleys. There is thus afforded a very strong contrast between the sharp peaks and needles of the Needle Mountains and the comparatively low relief of the reéxposed and ancient land surface extending southward from them.’ Of great topographic importance was the formation (Tertiary) of three series of volcanic accumulations: (1) a tuff which attained a thick- ness of 3000 feet east of Ouray (San Juan), (2) lava flows, tufis, and agglomerates (Silverton Series), also reaching a thickness of about half 1 Cross and Spencer, La Plata Folio U. S. Geol. Surv. No. 60, 1899, p. 8, col. 1. 2 Idem, p. 10, col. 3. 3 Idem, p. 2, col. 1. 4 Cross and Howe, Needle Mountains Folio U. S. Geol. Surv. No. 131, 1905, p. 1, cols. 2-3. ROCKY MOUNTAINS. III 379 a mile in places; and (3) later thin lava flows and tuff of a third period of volcanic activity (Potosi Series). In connection with the formation of these materials there were intruded into the sedimentary rocks various dikes, sills, laccoliths, and irregular masses which are the deeper-seated equivalents of the surface volcanics. On the flanks of the Needle Mountains dome a simple consequent drainage was developed during the doming period. The southern flanks Fig. 129. — Mount Wilson group in background. Looking down Howard Creek from south of Ophir Pass. of the dome were not covered with volcanic material and the conse- quent drainage developed on these slopes was not affected by volcanic action. The northern slopes of the domes, on the other hand, were affected by volcanic outpourings and the drainage greatly modified. After the last great eruptions the streams proceeded to dissect the surface deeply. The consequence of further uplift of the region and resulting deep dissection has been the formation of a confusing variety of slopes in intimate relation to the detailed geologic structure. Within short distances of each other one may find (1) ancient granites, (2) sedi- mentaries of more recent origin, (3) Tertiary conglomerates, and (4) more recent volcanic accumulations. The development of slopes upon these varied rocks has naturally been very complex and few generalizations may be applied to individual mountains or to the group as a whole. 2 Grenadier Range on west, White Dome in center Fig. 130.— Part of Needle Mountains, Colorado. ith each other. lews are panoramic W These two v about the head of Vallecito Creek. 380 ROCKY MOUNTAINS. III 381 GLACIAL FEATURES A phase of the recent dissection of the region is associated with gla- ciation; the conditions prevailing in the Needle Mountains during the portion of the glacial period of which the best records are left were nearly the same as those which exist in the Alps at the present time. From the positions of well-defined terminal and lateral moraines it is inferred that during the last recognized period of glaciation alpine glaciers descended the more favorably located valleys some 25 miles. The rock débris in the terminal moraines consists largely of material derived from the Needle Mountains. The higher and sharper peaks and ridges were not buried beneath ice and snow, but at lower elevations the bare rock surfaces are grooved and polished and indicate vigorous glacial action. ELK MOUNTAINS The principal range of the Elk Mountains lies north of the San Juan region and west of the northern end of the Sawatch Range. The moun- tains have great diversity of color, due to the presence of light-gray trachyte, red, maroon, and brown sandstone, etc. They are from 13,000 to 14,000 feet high, and are characterized by sharp, conical peaks and ragged, serrated ridges, pinnacles, and spires. The main range is bor- dered throughout a large part of its extent by broad, high, flat-topped spurs or secondary ranges.' It is about 4o miles long and dissected by gorges and canyons whose extreme ruggedness and picturesqueness are owing to the complicated structure, the variegated rocks, and the youthfulness of the forms themselves. Enormous amphitheaters at the heads of the bordering valleys have been cut back so far as to give the main crest a zigzag course and to make it so narrow as to be almost impassable.” PARKS OF THE SOUTHERN ROCKIES Between the two main ranges of the Rockies in Colorado is a line of basins of exceptional interest. The three northernmost, North, Middle, and South parks, are geologically a unit. Their southern continuation is not San Luis Park, as generally supposed, but Wet Mountain Valley and Huerfano Park between the front range axis on the one hand and the Sangre de Cristo axis on the other. The San Luis basin lies between the latter axis and the Sawatch Mountains. The former depression began to take shape at least as far back as the Triassic; the San Luis Valley is occupied chiefly by late Tertiary and Pleistocene sediments.® 1 H. Gannett, U. S. Geol. and Geog. Surv. of Colorado and Adj. Terr. (Hayden Surveys), 1874, Pp. 417. 2 Hayden Surveys, 1874, pp. 58, 71. 3 C. E. Siebenthal, The San Luis Valley, Colorado, Science, n. s., vol. 31, 1910, p. 745. 382 FOREST PHYSIOGRAPHY San Luis Park is the largest of the five main parks of Colorado. It is in many respects like the Sage Plains of the Green River Valley or the valley of the San Joaquin in California, consisting of a depression floored by young sedimentary strata and still younger alluvium; the lost rivers on the piedmont alluvial plain flanking the mountainous margin of the basin are also characteristic. Each of these parks drains to the sea by way of a stream which cuts through a deep mountain gorge on the margin of the basin. North Park is drained by the Platte, South Park by the South Platte, Middle Park by the Grand, and in the mountains enclosing San Luis Park the southward-flowing Rio Grande takes its rise. These parks are all the more striking because they lie in the midst of a very rugged mountain region. Their unique character is due primarily to broad intermont structural depressions whose low relative elevation and high outlets have made them the seat of aggradation, not dissec- tion, while the neighboring uplifts are among the loftiest ranges in the country and have been profoundly dissected, giving rise to a rugged and, locally, even an alpine topography. With these general relations in mind we shall now note briefly the special features of each park. NORTH PARK! North Park, 7500 feet above the sea, is nearly quadrangular in shape; it is about 50 miles in extent from east to west and 30 miles across from north to south. Its surface, although somewhat rugged, is marked by broad bottoms along the streams, especially the North Platte and its tributaries. The lofty and, for a part of the year, snow-covered moun- tains enclosing it like a gigantic wall, are covered with a dense growth of pine, but scarcely a single tree may be found over the whole 1500 square miles of the park itself save along the stream courses. Grass is found in abundance, and the park was originally the feeding ground of thousands of antelopes. The bordering strata dip under the park, becom- ing nearly or quite horizontal toward its center. In this respect North Park resembles the three other major parks of Colorado in being pri- marily a great structural depression sufficiently free from mountain- ous elevations and hence sufficiently drier to form a striking contrast to the high forest-clad country about it. MIDDLE PARK Although Middle Park has on the whole a basin-like aspect, yet in detail this feature is often lost in the prominence of many of the ridges separating its secondary drainage lines. In this respect it is in sharp 1 F. V. Hayden, U.S. Geol. Surv. of the Terrs., 1867, 1868, and 1869, pp. 87-88. ROCKY MOUNTAINS. III 383 contrast to both North and South parks on either hand. The basin is unique in being the easternmost region in the United States where Pacific waters take their rise; it drains westward through the Grand River to the Pacific. The borders of the basin are composed of crystalline schists and granites on the eastern, southern, and western sides, while the northern border as well as much of the floor of the basin is com- posed of younger sedimentaries.! d SOUTH PARK South Park is about 45 miles long and about 40 miles wide at its widest portion near the southern end. Its elevation at the north is about 9500 feet, at the south about 8000 feet. Its surface is regular only by contrast with the more rugged country surrounding it. Numerous parallel ridges with northerly trend cross the park and make its surface irregular. At the southern end are many isolated buttes, most of them of volcanic origin. Portions of the park are underlain by rather flat- lying sandstone, as is the case with San Luis Park, and such portions have as a rule a more uniform surface than the rest.’ SAN LUIS PARK The San Luis Valley basin owes its origin primarily to the broad, gentle, trough-like depression of its underlying sandstones. In late geologic time it probably contained a lake of considerable extent, a SIERRA CONEJOS MOUNTAINS LIMIT OF FLOWING WELL8 S-A:N), DUS VA ELEY VQLIMIT OF FLOWING WELLS Santa Fe formation Alamosa=tocma tion AA $1] 7000 = or SS Fy 23H 6000 EH eds, gravel, Lava Lava Santa fe formation conglomerates, etc. Approximate scale (no lava) Ky 0 5 Miles Fig. 131. — Cross section of San Luis Valley from foot of Blanca Peak to foot of Conejos Range. (Adapted from Siebenthal, U. S. Geol. Surv.) condition inferred from the character of the fine sediments accumulated in it2 The draining of the lake by the cutting down of the outlet stream, the building up of great alluvial fans, among which that of the Rio Grande is the largest, and the deposition of glacial deposits locally about its border are still later occurrences of physiographic importance. 1 Emmons, Cross, and Eldridge, Geology of the Denver Basin in Colorado, Mon. U. S. Geol. Surv. No. 27, 1896, p. 4. 2 A.C. Peale, U. S. Geol. and Geog. Surv. of Col. and Adj. Terr. (Hayden Surveys), 1873, p- 212. 3 F. M. Endlich, U. S. Geol. and Geog. Surv. of Col. and Adj. Terr. (Hayden Surveys), 1873, DP. 334, 339- 384 FOREST PHYSIOGRAPHY San Luis Valley appears nearly level over great areas but in fact it departs from the horizontal by important amounts. The marginal slopes descend by regular gradients to an axial depression located well toward the eastern margin of the valley; they are developed upon a series of coalescing alluvial fans and the longest and flattest slopes are those built by the largest streams; the Rio Grande has built an alluvial fan so large as to throw the axis of the valley to one side (east) of the center of the depression.’ The eastern side of the basin is. bordered by Fig. 132. — Looking eastward from Hunt Springs across the north end of San Luis Valley; San Luis Creek in the middle distance, Sangre de Cristo Range in the background. (Siebenthal, U. S. Geol. Surv.) the Sangre de Cristo Range, on whose flanks a great alluvial plain has been formed whose deposits above an elevation of gooo to soo feet consist of fluvio-glacial débris formed in connection with Pleistocene glaciers. Concentric terminal moraines surmount the crests of the alluvial fans and cones, Fig. 131.7 1 C. E. Siebenthal, Geology and Water Resources of the San Luis Valley, Col., Water- Supply Paper U. S. Geol. Surv. No. 240, 1910, p. 10. 2 C. E. Siebenthal, Notes on Glaciation in the Sangre de Cristo Range, Colorado, Jour. Geol., vol. 15, 1907, p. 15. Idem, The San Luis Valley, Colorado, Science, n. s., vol. 31, 1910, p. 740. ROCKY: MOUNTAINS. III 385 Northeastward from Antonito there stretches a line of flat-topped to rounded basaltic hills which represent late flows upon the alluvial floor of the valley. They form the chief topo- graphic break in the continuity of the basin floor. The smooth valley surface is also inter- rupted in a number of localities by sand dunes. The highest occur between Medano and Sand Creek and these also cover the most extensive single tract, 40 square miles. They con- sist of rather coarse, white, quartzite sand blown into place by the heavy southwest winds that occasionally blow for a two or three day period across the valley. Elsewhere the sand is com- monly blown out of place between patches or clumps of bush and grass, leaving small hollows or basins that contain lakelets in the wet season.1 OTHER TYPES OF PARKS Besides these large intermont parks there are scores of smaller ones. They have been formed in at least four main ways. (1) San Luis Park, as we have seen, is a broad structural basin drained by the Rio Grande River, and the same is true of North, South, and Middle parks. (2) In the Pikes Peak region and specifically in the south- central portion of the Pikes Peak quadrangle, is a depression known as Shaw’s Park; it is formed upon sandstones, marls, and limestones in the form of a syncline whose margins have been in a number of cases down- faulted in a pronounced manner. The weathering of the softer sedi- mentaries has accentuated the flatness of the depression normal to the synclinal structure and thus developed a lowland 5 or 6 miles in width. A similar park has been developed under similar structural conditions immediately west of Shaw’s Park and is known as Twelve Mile Park. (3) A third type of park may be seen very commonly throughout the region, a typical occurrence being Low Park, shown in the Needle Moun- tains topographic sheet, in the valley of the Florida River, where a broad valley lowland is formed by combined glacial erosion and the present drainage.? Similar valley lowlands of small extent are not un- common at the junction of tributary and master streams and are com- monly known as “‘parks.’’ (4) A fourth type of park is described by Powell,? who calls attention to the fact that the great anticlinal folds in the Park Mountains have a north-south trend and that on the flanks of each fold there is developed a zone of maximum dip so that the rocks at the flanks of the ranges are turned abruptly down. In such cases small parks are formed by the wearing out of the softer beds, leaving 1 C. E. Siebenthal, Geology and Water Resources of the San Luis Valley, Col., Water- Supply Paper U.S. Geol. Surv. No. 240, toro, pp. 47-48. The detailed features of the San Luis basin are represented in Sheet 1o, Atlas of Colorado, Hayden Surveys, where the course of the Rio Grande is shown as well as the course of San Luis, San Isabel, and other creeks which empty into San Luis Lake instead of having a surface channel to the Rio Grande. 2 Pikes Peak Folio U. S. Geol. Surv. No. 7, 1804. 3 Physiography of the United States, Nat. Geog. Soc. Monographs, 1896, pp. 88-80. 386 FOREST PHYSIOGRAPHY the harder strata standing as great steep walls parallel to the axes of the principal ranges. The famous Garden of the Gods near Pikes Peak is explained in this way. TREE ZONES AND TYPES Among the smaller parks Estes Park shows a typical zonal arrange- ment of the tree species of northern Colorado. There are three well- defined forest types: (1) the yellow pine type, developed typically as an open woodland which reaches from the upper foothills to an altitude of gooo feet; it occupies the ridges and slopes of the lower part of the park, but at higher elevations occurs only in the open valleys. In the open it occurs on small rocky knolls and ridges, whence it invades the grasslands of the park floor. (2) The lodgepole pine- Douglas spruce type occurs above the yellow pine type between 8000 feet and 10,000 feet, though below the lesser altitude the growth is mixed to some extent with the yellow pine, while above the greater altitude it is mixed with Engelmann spruce. (3) The Engelmann spruce-alpine fir type constitutes the forest at timber line and usually some distance below it. Timber line is here at 10,500 feet. From 8500 to gooo feet the grass land of Estes Park is in the form of typical mountain meadows.! Among the larger parks San Luis exhibits a rather typical distribu- tion of vegetation. On the high mountain sides flanking San Luis Valley are pine, aspen, and spruce. Lower down are pifon and cedar; in the valleys and along the streams are cottonwood and willow. Away from the streams and on the arid basin floor greasewood is the principal growth.” 1 F. E. Clements, The Life History of Lodgepole Pine Forests, Bull. U. S. Forest Service No. 79, 1910, pp. 7-8. 2 C. E. Siebenthal, Geology and Water Resources of the San Luis Valley, Col., Water- Supply Paper U. S. Geol. Surv. No. 240, 1910, p. 26. CHAPTER XXI TRANS-PECOS HIGHLANDS MOouNTAINS AND BASINS THE Trans-Pecos region embraces an assemblage of topographic forms which individually resemble the features of adjacent provinces, the Rocky Mountain province on the north, the Arizona Highlands province on the west, and the Mexican plateau on the south, and it may therefore be regarded as a transition province. The Trans-Pecos ranges do not have that continuity which marks the main mountain ranges of the Pacific Cordillera. They exhibit a great variety of slopes produced at different geologic times and in many different ways; the province includes a mixed group of topographic forms not conveniently classified with the forms of adjoining provinces. In this respect it is unlike the other physiographic regions of the United States, each of which has a certain essential unity of structure or topography or both. Mountains and intermontane plains of several types and of variable extent are the predominating features of the Trans-Pecos country. The latter are in part plains of degradation, in part broad constructional plains built up by detritus from the bordering mountains and called bolsons (Spanish bolsén, for “large purse”). The trend of the moun- tains and intervening basins is north-south, except near and south of the New Mexico-Texas boundary where they trend northwest-southeast. In general the elevations are distinctly lower than those of any other portion of the Pacific Cordillera in this longitude. Only two peaks rise above 8000 feet and the general elevation of the lowlands is from 3500 to 4500 feet. While the greater number of the Trans-Pecos mountains have a fault- block origin, as a whole they are of diverse origin, structure, and topog- raphy. They exhibit many irregular forms of relief and all have sharp and rugged outlines. Few of them rise to the height of the dry timber line (6000 feet). On the summits and in the higher canyons of the Sacramento, Chisos, and Davis mountains, and a few others, this ele- vation is exceeded and a certain amount of timber is found. On the lower and drier mountains the vegetation consists of edible pines, a variety of junipers, and several species of maguey. The lower eleva- 387 388 | FOREST PHYSIOGRAPHY Hayy, C "DEAF SMITHO sk Laplata® ty a% “Dimrmitty MER? CASTRO. I 4 ~ Talia he ee SWISHER ATCHISON PART OF TEXAS AND ADJOINING TERRITORIES Scale of Miles 75 100 Grande Fig. 133. — Principal mountain ranges of the Trans-Pecos province, chief drainage features, and the bordering Llano Estacado on the east. (U. S. Geol. Surv.) OF TRANS=PECOS HIGHLANDS 389 tion of the group as a whole is expressed in the lesser rainfall and general barrenness of the mountains in contrast to the partly timbered Rocky Mountains to the north. The highest summit is Sierra Blanca (1,880 feet) in the Sacramento range in southern New Mexico, but in general the peaks fall short of this altitude. s MountTAIN TYPES Four types of mountains may be identified: (1) mountains of deforma- tion, consisting of structural folds or tilted fault blocks in which the outline of the mountain is in sympathy with the structural breaks or Fig. 134. — Western face of the Fra Cristobal Mountains, New Mexico. Two faults, a and b, give prom- inence to the mountain front. They are represented in the section below the photograph. (Lee and Girty, U. S. Geol. Surv.) deformations, for example the Sandia and Manzano mountains east of Albuquerque; (2) plateau mountains, consisting of nearly horizontal plateaus without important deformations. This type of mountain occurs in the form of summits or shoulders and in close proximity to 390 FOREST PHYSIOGRAPHY the higher relief features of the province.!. (3) Mountains due to the upthrust of a granitic core through sedimentary rock that now forms the borders of the range, as the Black and Mimbres ranges. (4) Moun- tains consisting chiefly of volcanic material like San Mateo-and Valles Pee RUN TS. HUECO MTNS. : a Sane SEMIN Se eas chee er ab Ch os : AL = RE rere a SSS = se Seal RIO GRANDE Fig. 135.— East-west section across the Trans-Pecos Highlands north of El Paso, Texas. pC, pre-Cambrian quartzite and porphyry; COS, Ch, gs, etc., paleozoic limestone, sandstone and gypsum; K, cretaceous limestone and sandstone; gr, intrusive granite; Qb, Quaternary basin deposits. Scale, 1 inch repre- sents nearly 50 miles. (Richardson, U. S. Geol. Surv.) mountains and others. The members of the last-named group fall into three subtypes: (a) old igneous vents such as dikes and necks, (b) vol- canic craters, and (c) mesas capped by sheets of lava that have pro- tected the underlying rocks from the denudation that carried away the surrounding material. The Trans-Pecos province is one of the least-known physiographic provinces in the United States and few generalizations may be made concerning details of mountain form beyond those in the preceding paragraph. It is known that the province as a whole is separated from the Rocky Mountains province at Bernal, New Mexico, by a broad level plain or plateau, and it is not until about too miles south of Bernal that the mountains of the Trans-Pecos begin in the Jicarillas, the north- ern outliers of a chain of mountains, the loftiest ranges of the province, that extends with marked continuity through New. Mexico and Texas as far as the 32d parallel. The individual members of the chain are the Jicarillas on the north, and next in order from north to south are the Sierra Blanca, Sacramento, Sierra Guadalupe, Comanches, Caballos, and the Sierra de Santiago. These constitute the front ranges of the province and they give a distinctive character to the eastern edge of the highland belt in which they lie. The rocks composing these mountains are chiefly limestone and sandstone. The general structure of the east- ern border ranges is an eastward-dipping monocline ending at the Rio Pecos, but there are many lateral ridges and separate peaks. On the west the mountains of the eastern border terminate in a steep scarp 1000 to 2000 feet high. The long eastern slopes correspond with the dip of the underlying strata. West of the front ranges of the Trans-Pecos country are other ranges of the fault-block type. They occur as short sierras or groups of sierras in the form of isolated mountains or elongated chains surrounded by extensive and nearly level desert plains. They have lower altitudes 1 R.T. Hill, Physical Geography of the Texas Region, Folio U.S. Geol. Surv. Ne. 3, 1900, p. 3. TRANS=PECOS HIGHLANDS 391 than the front ranges, and so isolated are the various members that no group name has been given them. The principal ranges are, however, arranged in two lines on either side of the Rio Grande Valley. On the east side are the Sandia, Manzano, Oscura, San Andreas, and Organ ranges; on the west are the Nacimiento, Limitar, Magdalena, Cristobal, Fig. 136. — Fault-block mountain in the Trans-Pecos Highlands, Texas. (Marfa quadrangle, U. S. Geol. Surv.) Caballos, and Cuchillo Negro ranges. The fault scarps of these ranges are always steep and in general face inward toward the Rio Grande Valley, which is thus a great structural depression or series of elongate basins, though there are some important exceptions to this rule. They are all high and in process of vigorous dissection, and hence display a s 392 FOREST PHYSIOGRAPHY varied relief. The trend of the group is, however, definite, due to their fault-block origin, and of similar definiteness are the valleys that occur among them. Westward of these block mountains are many short independent blocks surrounded by bolson deserts and not capable of union into related chains. They rise from desert plains and in general Fig. 137. — Western escarpment of the Caballos Mountains, New Mexico, and the dissected alluvial slope at their base. (Lee, U. S. Geol. Surv.) have steeper descents on their western than on their eastern sides. They consist for the most part of fault blocks with monoclinal dips, although in places extensive folded structures also occur. FRANKLIN RANGE The most detailed study of mountain form and structure in the Trans-Pecos region is by Richardson, whose structural sections of the Franklin range are shown in Fig. 138. They will serve as illustrations of a type of mountain found in the province. The Franklin range is a long, narrow, crust block resembling the mountains of the basin-range type in its general features but differing from many though not from all of them in having a more complex system of internal faults. The strata strike parallel to the trend of the range and dip westward at steep angles. On the steeper eastern face of the Franklin range are TRANS-PECOS HIGHLANDS 393 exposed the eroded edges of the strata composing the fault block, while the gentler westward slope coincides to an important degree with the dip of the underlying strata. The range is broken into several second- ary blocks by normal parallel faults some of which strike with the trend of the range, but there are also several transverse fractures, and the strike of a few of the faults is a curve. Both the eastern and the western margins of the range are marked by faults; that on the western border consists of two parallel faults at the base of the range between the foothills and the main mountain mass. The greatest displacement is in the central part of the range, where a throw of 2500 feet has been determined. Fig. 138. — Section across Franklin Mountains, showing relation of structure to relief. In, quartzite; thp, rhyolite porphyry; Cb, sandstone; Oep, Om, Sf, Ch, and Kcm, limestone; gr, intrusive granite; Tap, andesite porphyry; Qb, basin deposits. Scale 1 inch represents 2 miles. (Richardson, U. S. Geol. Surv.) Fig. 139. — East-west section across the Franklin Mountains north of El Paso, Texas. To supplement Fig. 138. Legend and scale as in Fig. 138. The fault at the eastern foot of the range is completely concealed by wash and its position is hypothetical. That the mountain front is the locus of a fault is suggested by the evidences of geologically late (probably Quaternary) faulting at El Paso, where displacements in allu- vial deposits may be seen, and by the benches west and northwest of Fort Bliss at eleva- tions ranging about 4ooo feet. The benches are the upper portions of broken alluvial slopes which in places fringe the base of the range in an uneven eastward-facing scarp varying from to to 50 feet in height. In many places the bench has been destroyed by erosion along the many arroyos that descend from the adjacent mountains. The original fault along the base of the range is probably of ancient origin, for the drainage has diversified the eastern border and to a large extent obliterated that inequality of slopes that obtains in the case of a young block mountain. VOLCANIC MOUNTAINS AND PLATEAUS? In addition to the mountains formed out of tilted crust blocks of sed- imentary material there are many eccentric and irregular mountains of volcanic origin that rise from broad plains. Of this type are the Chisos (ghosts), Corazones (hearts), the Sierra Blanca, and the Davis moun- tains, among which the last-named are the most extensive. They con- sist of a group of volcanic necks and dissected volcanic plateaus that end 1G. B. Richardson, El Paso Folio U. S. Geol. Surv. No. 166, 1909. 2 R. T. Hill, Physical Geography of Texas, Folio U.S. Geol. Surv. No. 3, 1900. FOREST PHYSIOGRAPHY “(orurereT) wUNyeAys BULAT-Jey prey v uo podojadrdgp st 99v149} JuIU -twoid sy, *punossai0y ayy ur ureyd oy} aAoqe yoo} o00f Suisr1 vsout poddo}-yey oy} pue “qyeseq Jo SHII [VIIWAOA SMOYS ‘PSOTAT UOJPY puv yrog $,doyst J — ‘ovr “aL TRANS=-PECOS HIGHLANDS 395 at the south in a great escarpment over 1000 feet in height that over- looks the continuation of a broad intermontane plain toward the south. The mesa and cuesta types of mountains are capped in some in- stances with old igneous material in places 500 feet thick. The volcanic caps are remnants of former sheets of similar material that have been all but removed by denudation. Elephant Cuesta and Nine Point Mesa are illustrations of this kind of mountain. Besides these are other volcanic mountains of recent origin, cinder cones or true volcanic craters accompanied by sheets of lava which flowed from them. The craters are of interest because they are the most easterly known in the United States and the only ones lying east of the front of the Cordillera. The cinder cones have been formed since the degradation of the Mesa de Maya and Ocate plateaus, for they rise out of the newer and lower plain below the latter. The most conspicuous crater is Mount Capu- lin, a very symmetrical cinder cone with a vast crater at the top of it. Within the Trans-Pecos Highlands are extensive areas of uplands with sublevel summits bordered more or less completely by pronounced escarpments. They occur as large benches that border the mountain ranges or that rise from the bolson deserts. They are either com- pletely isolated from the mountains or project bench-like from the bases of the mountains. They are lava-capped mesas whose summit layers of hard volcanic rock have protected the more friable material underneath them; they are distinguished from the flat-topped mountains noted above only by their greater extent and lesser elevation. They are cut by deep canyons, bordered by steep escarpments, and owe their present outlines to marginal erosion. Northeastern New Mexico, Fig. 133, is therefore not a part of the Great Plains but an eroded plateau of Cretaceous rock surmounted by basaltic flows.! Red Beds MESA DE MAYA | Dakota Sandstone | | Colorado Shale Laramie Shale Basalt | r—) Basalt Basalt LENGTH OF SECTION=100 MILES HEIGHT OF SECTION. = 2 MILES Fig. 141.— Upturned strata on the eastern border of the Rockies, the anticlinal structure of the bordering strata underneath the Mesa de Maya, and the protective influence of the basalt cover. (After Keyes.) The three most conspicuous plateau plains of northern New Mexico are known as the Mesa de Maya, the Ocate Mesa, and the Las Vegas 1 R. T. Hill, Notes on the Texas-New Mexico Region, Bull. Geol. Soc. Am., vol. 13, 1892, pp. 85-100. 3 96 FOREST PHYSIOGRAPHY Mesa. The western portion of Mesa de Maya is also known as Raton Mesa or Raton Mountains. The mesa is a long dissected plateau almost due east of Trinidad, Colorado, which extends southward into New Mexico and Texas. Its border rises nearly 4800 feet above the adja- cent valleys and the massive summit is composed of thick beds of old lava now dissected into remnants of a once more extensive plateau. Underneath the lava are less resistant sandstones and shales (Upper Cretaceous). Raton Mesa is 8 miles long, 4 miles wide, and includes about 20 square miles. It has been entirely separated by erosion from a similar area 103°%0° EER ND : aes | ; Aras . 10340 Fig. 142. — Topography of Mesa de Maya, south-central Colorado. Typical lava-capped mesa with steep bordering scarps and Flattish summit.. (U. S. Geol. Surv.) to the south and from the main lava field to the east. The aggregate thickness of the basalt flows which constitute the cap rock of Raton Mesa and of the Mesa de Maya is from 250 to 300 feet about their borders, and increases to 500 feet toward the central part of the western mesa. At least 8 distinct beds of lava have been identified, which prob- ably represent the same number of. independent eruptions. ‘The differ- TRANS=PECOS HIGHLANDS 397 ent beds are 30 or more feet in thickness, though they vary greatly from place to place! Raton Mesa has a mean elevation of about 1800 feet and is bordered by an almost vertical escarpment from 200 to 300 feet high, an escarpment so sheer as to render it almost inaccessible, Fig. 141. There is a dense growth of pifon and juniper along the base of the Raton Mesa; the small streams that head in the plateau are fringed with cottonwoods; on the steep border of the plateau, pine and spruce trees are scattered through a dense undergrowth of scrub oak, with aspen appearing locally near the base of the escarpment. The entire district up to 7500 feet affords a scanty growth of nutritious grasses well suited for sheep farming, which is one of the chief indus- tries. The tableland on the summit of the mesa supports a growth of bunch grass. The Ocate Mesa has a cap rock of thick sandstone and it is the rem- nant of a lower-lying mesa that extends westward to the foothills of the Rocky Mountains between the Arkansas River in Colorado and the Cimarron River in New Mexico. The largest development of the Ocate Mesa is between the Cimarron Valley and the Mora, where from a broad platform rises Ocate crater (8900 feet). The eastern border of Ocate Mesa is an escarpment nearly 500 feet above the plain to the east, known as the Las Vegas Mesa, which extends south of Mesa de Maya to the great cliffs of the Canadian valley and east toward the Great Plains near the Texas line. The surface of Las Vegas Mesa slopes gently to the east; the western border has an average altitude of about 8600 feet. It is a vast stratum plain underlain by a chalky sandstone (Colorado formation). Many volcanic craters and dikes rise out of it to diversify and complicate its relief. The cliffs which terminate the eastern border of Las Vegas Mesa form one of the longest and most remarkable escarpments in America. They are continuous with the cliffs of Galistes Mesa and extend in an indirect manner from the ro4th meridian southwest to the 106th meridian, or nearly 300 miles. The cliffs are developed upon sandstone and overlook the deep Pecos Valley. DRAINAGE FEATURES A detailed map of New Mexico and western Texas, Fig. 142, displays many features characteristic of surface drainage in arid lands. A large number of enclosed basins of structural origin occur, and these are in the main floored with loose sediments derived from the surrounding moun- tains. About the margins of the basins are long talus slopes, huge bowlder and alluvial fans, and dry arroyos. The basins are without 1 R. C. Hills, Elmoro Folio, U. S. Geol. Surv. No. 58, 1899, p. 3, col. 2. 2 Idem, p. 1, col. 1. 308 - FOREST PHYSIOGRAPHY surface streams of any extent, a few watercourses entering them from surrounding highlands, but the porous nature of the soil, the high rate of evaporation, and the low rainfall, cause the streams to disappear before they have traveled far from the margins of the basins. The annual rainfall is in some places less than 10 inches, over the greater part of the region it does not exceed 15 inches, and everywhere it is chiefly in the form of summer thunder showers of short duration and limited extent. All of the basins or bolsons, as they are sometimes called, are charac- terized by ‘‘lost rivers,’’— streams that disappear about the borders of the basins. The floors of some of the basins are occupied by salt marshes or temporary lakes. In the largest basins, these between the Sierra Madre and Guadalupe mountains, the salt deposits are of great extent and have been used for hundreds of years by the Mexican population as one source of a salt supply. Around the margins of some of the bolsons are benches consisting of fan-shaped heaps of land waste de- rived from the mountains and deposited by torrential streams. The word “bolson” is of Spanish origin and according to Hill it designates structural valleys between mountains or plateau plains which have been partially filled with débris from adjacent eminences.!. Bolson originally meant a constructional plain bordered by mountains or plateau escarpments that supplied the material with which the bolson is floored; the term did not cover the bordering mountains, though the definition implied that the bordering moun- tains are features normally associated with the plain, and that a bolson plain was part of an interior basin.2. In short, a bolson plain is the floor of an interior basin with such variations in its expression as are naturally associated with variations in climate, disposition of waste, size of basin, length of existence, etc. The basin floors are characteristically flat and in broad view are somewhat similar in appearance to the plateau plains just described. They are, however, to be distinguished from the plateau plains by the fact that they are, at least in part, surfaces of aggradation, while the plateau plains are on the whole either destructional surfaces or surfaces that have been at least partly preserved from destruction by a cap of resistant lava. LONGITUDINAL BASINS The basins of the Trans-Pecos region occur in four longitudinal belts in sympathy with the belted arrangement of the intervening mountains, Fig. 133. The easternmost lies along the eastern fronts of the Guada- lupe and Santiago ranges. The second belt lies between these ranges and the Sierra Diablo and Cornudas mountains. The third line of basins lies between the Sacramento and Sierra Blanca mountains on the east and the Sierra Oscura and San Andreas ranges on the west. 1 R. T. Hill, The Physical Geography of the Texas Region, Folio U. S. Geol. Surv. No. 3, IgOO. 2 W. G. Tight, Am. Geol., vol. 36, 1905, pp. 271-284. TRANS-PECOS HIGHLANDS 399 The fourth line lies between the last-named mountains and the Sierra de los Caballos on the borders of the Rio Grande. Among these four longitudinal basins the Hueco basin, Fig. 133, is the largest. It lies partly in New Mexico, partly in Texas, is about 200 miles long, about 25 miles wide, and stands about 4000 feet above the sea. The mountains that border and enclose it are 2000 to 5000 feet higher. A few miles north of the New Mexico-Texas boundary the basin is crossed from west to east by a low divide. The southern half is drained by the Rio Grande; the northern half is a closed basin with salt marshes and unusually interesting dunes of white gypsiferous sands, a district known as the Tularosa Desert. In the vicinity of El Paso the Hueco basin is a structural basin or trough deeply filled with detritus. The streams wither at the foot of the mountains, and the mouths of the mountain arroyos are marked by huge detrital cones and fans that spread radially outward and finally coalesce with the wash from the intervening slopes. JORNADO DEL MUERTO The most noted basin of the Rio Grande Valley is the Jornado del Muerto Bolson. Its name means “the journey of death” and was given to it in pioneer days because of the great difficuity then involved in crossing its dry, barren, and inhospitable wastes. The Jornado basin lies in south-central New Mexico. It extends north and south a distance of more than 200 miles, and is from 30 to 40 miles wide. It is a flat-bottomed basin except along the Rio Grande where the river has cut a valley 400 feet below the general level. Its margins are upturned steeply on all sides to heights of 2000 and 3000 feet above the general level. In the central part of the basin are a number of shallow depressions which hold storm waters that some- times linger as lakes for several months, seldom through the year. At various points the even surface of the basin floor is broken by low hills of volcanic origin, among which may be mentioned the Dona Ana Hills, San Diego Mountains, and Cerro Roblero in the southern part of the basin. All of the volcanic cones are of recent origin and some of them have perfectly preserved craters. From some of the cones basalt flows have extended out over the surrounding plain, the one south of San Marcial covering more than 100 square miles. It has been shown recently! that certain interior basins of the Trans-Pecos country, among them the Jornadc del Muerto, instead of being structural valleys deeply floored by mountain waste, have a rock surface developed on the beveled edges of the strata, representing a total 1 C. R. Keyes, Rock Floor of Intermont Plains of the Arid Regions, Bull. Geol. Soc. Am., vol. 19, 1908, pp. 63-92. 400 FOREST PHYSIOGRAPHY thickness of thousands of feet, and that for the most part the rock floors of these plains are covered by a thin veneer of detritus only instead of the thick alluvium usually ascribed to them. Rarely does the thickness of the detrital mantle exceed 100 feet. The general slope of the Jornado del Muerto plain for example is only 2° or 3°, while the dips of the strata are in many cases as high as 30° in the same direction, and even vertical. On the beveled edges of these steeply inclined beds alluvium and also broad sheets of basaltic lavas have been laid down. BASINS OF THE RIO GRANDE VALLEY Among the most important basins of the Trans-Pecos country are those drained by the Rio Grande. This river from the San Luis Valley in southern Colorado to the point where it cuts the easternmost mountains of the province (long. 103° W.) flows almost continuously Fig. 143. — Valley of Rio Grande, El Paso, Texas, showing passage of the river from the Mesilla Valley to the Hueco basin across the southern end of the Franklin range. Note the terraced margins of the dissected basins. (Hill, U. S. Geol. Surv.) through a series of old structural basins connected by canyons that increase in length and depth toward the southeast. The present course of the Rio Grande is owing in part to the basin feature and in part to volcanic action, and the activities of the river itself. Long after the partial filling of the broad structural troughs of the region there was an early period of valley cutting which was followed by a second period of valley filling, probably on account of changed topographic and climatic conditions. Near the close of the second period of aggradation great sheets of basalt were extruded. During this time the stream courses were in many instances violently changed and the Rio Grande, which formerly ran southward through the Jornado del Muerto basin into the interior basins of northern Mexico, was diverted into Engle and Las Palomas valleys, which are much narrower than the first-named basins. At the same time topographic changes due to volcanic action farther south probably forced the Rio Grande eastward to its present course south of the Franklin range at Hill Pass and south- eastward to the Gulf. The volcanic eruptions and changes of river courses were followed by TRANS-PECOS HIGHLANDS AOI a second period of erosion in which were formed the present narrow elongated valleys that stand at an elevation several hundred feet lower than the basins in which they were cut. A last phase of river activity is expressed in the form of silt accumulations on the floors of the flood plains, a filling with a maximum depth of 85 feet in the El Paso canyon. The river is carrying an immense quantity of silt to-day; on the average it transports about 14,580 acre feet of mud a4 year. The flood plain of the Rio Grande is cut well below the level of the Jornado del Muerto and is known in part as the Mesilla Valley. This division is about 45 miles long and 5 miles wide. Along the valley mar- gins terraces of notable extent and height have been formed. Some of them are of structural origin; their upper surfaces correspond with the dip of the rock in many instances, and rock outcrops along their valley- ward margin. Ata lower level are alluvial terraces representing com- plexities in the down-cutting of the Rio Grande since gaining its present course across the basins. The Rio Grande is a storm-water stream subject to great and sudden floods. The rainfall occurs principally in the form of violent showers or cloud-bursts which fill the dry stream beds with turbulent floods of short duration. When such floods occur simultaneously at many points they are likely to cause destructive floods on the valley floor where the fertile irrigable lands are located and where most of the population and the principal towns are to be found. CLIMATE, SOIL, AND VEGETATION The rainfall of New Mexico varies from 20 to 25 inches in the moun- tains, as above Las Vegas and Cloudcroft, to about 15 inches on the Great Plains to the east in the vicinity of Roswell and Carlsbad, so that the Pecos River, forming the eastern boundary of the Trans-Pecos Highlands, receives practically no tributaries from the east. The effects of the small rainfall of lower elevations are reénforced by the porous nature of the soil, upon which there is no extensive surface drainage; the water also drains rapidly away underground in many places through fissured limestone rock.2, At El Paso the mean annual precipitation is 9.85 inches, ranging between a maximum of 18.30 inches in 1884 and a minimum of 2.22 inches in 1891. The average humidity is 38.8 %, ranging between 23.2 % in May and 47.3 %in January.2 The mean annual precipitation throughout western Texas is about 12 inches. It falls mainly in the summer months in the form of brief showers and is variable and uncertain. 1 W. T. Lee, Water-Supply Paper U. S. Geol. Surv. No. 188, 1907, pp. 20-24. 2 Freeman, Lamb, and Bolster, Surface Water Supply of the United States, 1907-08, pt. 8, Western Gulf of Mexico, Water-Supply Paper U. S. Geol. Surv. No. 248, 1910, p. 114. 3 G. B. Richardson, El Paso Folio U. S. Geol. Surv. No. 166, 1909, p. 2. 402 FOREST PHYSIOGRAPHY In the lower Pecos Valley irrigation has reached a very high stage of development.1 Thou- sands of acres are under cultivation, beginning a short distance above Roswell and continuing into Texas; below this fertile belt but little irrigation is carried on because the seepage water contains a large amount of alkali, which renders it unfit for irrigation purposes. The floors of the arid basins in the Trans-Pecos region generally con- sist of fine detrital material and support a growth of stunted shrubs and grasses, such as mesquite, greasewood, and cactus. Along the river bottoms are cottonwoods and an undergrowth of shrubs. The coarser talus slopes are covered with a growth of yucca, cactus, and other desert flora and a scattered scrubby growth of oak, cedar, and juniper. These forms are gnarled and stunted at lower elevations, but in higher Fig. 144. — North end of San Mateo Mountains, Trans-Pecos Highlands, New Mexico. Shows char- acteristic tree-covered slopes. Mountains composed of rhyolite flows and tuffs. (Gordon, U. S. Geol. Surv.) situations they become increasingly more exuberant in growth.? Upon. the slopes and summits of the highest mountains, the Chisos, Davis, Capitan, and Sacramento ranges, a thrifty tree growth of pine and fir is found; upon the lower mountains the tree growth becomes more scattered and upon the lowest ranges no true forests appear. The Black range bears a good growth of pine upon its upper slopes; the Magdalena range and San Mateo Mountains have poorer growths of 1 G. B. Richardson, El Paso Folio U. S. Geol. Surv. No. 166, 1909, p. I15. 2 C. E. Dutton, 6th Ann. Rept. U. S. Geol. Surv., 1885, p. 125. TRANS-PECOS HIGHLANDS 403 105110/ Jp Arroyo Sierra ee Seale of Miles SoCo, OO OOo 2 1 0 2 4 6 8 10 Contour interval 1000 feet {J Timberless, Grazing [=] Woodland ZZ Less than 2000 feet B.M. ES 2000 to 5000 feet B.M. = H--45000 to 25,000 feet B.M. Fig. 145A. — Timber belts, Capitan Mountains, New Mexico. Note (1) the manner in which the timber belts follow the waterways, (2) the increase in growth with increase of elevation, and (3) the island- like outlines of the densest growths. (Adapted from U. S. Geol. Surv.) pine interspersed with cedar and juniper, though in the better-favored situations good stands are found.! The ranges of the species of trees occurring in the Trans-Pecos region vary as to rainfall, soil, and slope exposure. The lowest and best growth is found on the northern cooler and moister slopes and wherever the best soils have been developed. Except on the broader summits the soil has little humus for the undergrowth is scattered and light, oxidation of decaying vegetation is rapid and fairly complete, and erosion is in general active. There is a notable banding of the various species of forest trees, though the ranges of the different species overlap, as shown in Fig. 145B, which illustrates distributions in the Lincoln Forest Reserve of New Mexico (Capitan and Sierra Blanca ranges). The highest, or subalpine zone is found in this district between gooo and 11,000 feet. Engelmann spruce is the principal tree of this zone with subordinate amounts of white fir, red fir, Mexican white pine, and aspen, generally in groves. The yellow pine zone between 6400 and gooo feet includes, beside the dominant yellow pine, red fir, white fir, Mexican white pine, and oak. Between 5000 and 6400 feet is the woodland zone in which the principal species are pinon, juniper, cedar, scrub oak, and, along the streams, ash, box elder, and walnut. The Trans-Pecos forests seldom form dense stands; open scattered forest growth is the prevailing type. Since the forests grow only at the higher elevations, they are in relatively inaccessible situations, for the settlements are found in the valleys. The principal value of the forests 1 Lindgren, Groton, and Gordon, The Ore Deposits of New Mexico, Prof. Paper U. S. Geol. Surv. No. 68, 1910, p. 217. 404 FOREST PHYSIOGRAPHY is in relation to water conservation. Many small perpetual streams are limited to the forested zone; only a few advance even two or three miles into the bordering deserts. Forage grasses form a ground cover of great extent both in the parks of the forested zone and. below the BOXELDER } SCRUB OAK ENGELMANN SPRUCE WOODLAND YELLOW PINE SUBALPINE ZONE ZONE ZONE Fig. 145B. — Range and development of tree species in Lincoln National Forest, Trans-Pecos province. (U.S. Geol. Surv.) 7000 foot level where the forest ends. In the more accessible situations these have been irreparably damaged by overgrazing, for with the de- struction of the grasses deep and extensive erosion has taken place and rich pastures have been laid waste. 1 Plummer and Gowsel!, Forest Conditions in the Lincoln Forest Reserve, New Mexico, Prof. Paper U. S. Geol. Surv. No. 33, 1904, pp. 10, 11, 18. CHAPTER XXII GREAT PLAINS TOPOGRAPHY AND STRUCTURE THE Great Plains province slopes eastward from the foot of the Rocky Mountains to the valley of the Mississippi, where it merges into the Prairie Plains on the north and the Coastal Plains on the south. The Great Plains appear as wide areas of relatively flat tabular sur- faces crossed by broad shallow valleys drained by scores of eastward- flowing streams tributary to the Mississippi. Certain portions, as in central Montana, are drained by streams sunk in narrow canyons several hundred feet deep. The general expanse of smooth surface is also broken in some places by buttes, mesas, domal uplifts, extended escarpments, and local areas of ‘“‘badlands.”” In some districts ex- tensive areas are covered with sand hills, as in northwestern Nebraska, where is found a typical area of this character several thousand square miles in extent. As a whole, the province descends toward the east about to feet per mile and from altitudes of about 6000 feet on the west to about 1000 feet on the east, though the elevations and the general regional slope have considerable variation from place to place. An important illustration of variation in altitude is Pine Ridge, an irregular escarpment which extends from the northern end of the Laramie Mountains in Wyoming eastward to the northwestern corner of Nebraska and the southern part of South Dakota. It marks the northern boundary of the higher portions of the Great Plains, and from it cliffs and steep slopes descend northward 1000 feet into the basin of the Cheyenne River. North of the Cheyenne River the divides are much lower and the surface as a whole does not attain the level of the High Plains to the south. The plains topography is developed on a great mass of (1) rather soft deposits — sands, clays, and loams— spread out in the form of thin and extensive beds that slope gently eastward and (2) gently inclined and more indurated stratified rock varying from limestone and gypsum through shale and sandstone to conglomerate. The material of the first class is found chiefly in that subdivision of the Great Plains called the High Plains which extend from the southern margin of the bad- 495 406 FOREST PHYSIOGRAPHY lands to the parallel of 32° south in central-western Texas, and east and west extends between the rooth and to4th meridians. The hard rock elsewhere underlying the Great Plains is for the most part thinly cloaked with rock waste, and its topography is in large part-responsive to structure save (a) where the effects of base-leveling are still topo- graphically expressed or (6) where glaciation has modified the surface ot (c) where local alluviation has concealed an earlier structural surface. The two classes of material forming the Great Plains were derived Fig. 146. — Geologic map of the Texas regions, showing the relations of the Great Plains formations to those of the surrounding provinces. 1, Older granite; 2, Paleozic and Mesozoic; 3, Cambrian-Silurian; 4, Carboniferous; 5, Permian; 6, Jurassic; 7, Lower Cretaceous; 8, Upper Cretaceous; 9, Nonmarine Tertiary; 10, Marine Eocene; rr, Coast Neocene; 12, Later igneous. (Hill, U. S. Geol. Surv.) mainly from the west and deposited in stratified condition upon flood plains and lake floors and in part on the sea floor. The region has suffered broad uplift and depression a number of times in its geologic history, but these deformations have been regional in character and have not deformed the sedimentary series in a complex manner except in local instances.! 1 Darton and Salisbury, Cloud Peak-Fort McKinney Folio U. S. Geol. Surv. No. 142, 1906, p. 2, col. 1. j 407 GREAT PLAINS *4SvayNOs 0} JSaMy]IOU WOT} JAMO] 9Y} ‘}S¥a 0} 4SIA\ WOIJ Spud}Xo UO!}das Joddn oy, ‘o}VUIMULITLIP-Iwes oIe YIOG ‘suUle]_ JeoI4H) 9} SSOIOR SuOTzOaS JeINjonAys pure s1ydeisodoy, — "LPI “SIT SaTIW G ” yt fqyoLLysa S31IW 09 SLN3SANd3YN HONI L ‘TV.LNOZINOH ajeos eyewixoiddy VASVYSAN YO VLOWVG HLNOS NYSLSVA ONY WHYLN3S VLOWVG HLNOS SNINOAM auoysawl| ‘ayeys 'Hje y 2 poe ee ae ee —E = aeasineis Hee Payepllosuoo Ajped ‘oy 'pues ‘jaaud Avera 7 ae Hany wnoseyy J “S'S pur 's'| ‘|eq ayeys Peeesteee. ee }S1YOS puke ajiuelo yr ayounog aypoge esi pue Bue) SaNV avg . Sisterasled SNIV1d 1V4Y5D is STITH YOy1d" ‘SLN = NUYOHDIE SSW 9 ” yy ot fWOLLYSA S3TIW 09 SJ.NASSYdaY HON! | “Vv.LNOZINOH ajeog eyewxoiddy SVSNVYH YO VASVYEAN TWYLNSAD GNV NYSLSSM OGVYO109 NYSLSVAHLYON Sulelg Udi aun yo SS a er Teg eS ee nee “Pelepljosuoo Ajyied 9,9 ‘pues wanes Arena y e oo j z eyeus ‘guoyjspues 9u0}spues *guoy}soull| Ppue 9u0}sow!| Dlozosay SNIVId tLV3Y9 a ee SNIVId HOIH oavyo109 408 FOREST PHYSIOGRAPHY Although the strata of the plains lie sensibly flat within short dis- tances, the structure of the hard rock, as shown in Fig. 147, has the general character of a great geosyncline which rises on approach to the eastern border of the Rocky Mountains, where the steeply inclined, eastward-dipping strata appear in the form of hogback ridges, fringing the front ranges of the Rockies. This regional westward dip of the Great Plains strata is distinctly characteristic; but the general structure is modified in places by structural deformations of a certain degree of importance. In southeastern Colorado, for example, there is an arch in the form of an anticline which extends southeastward from the Green- horn and Wet Mountain ranges and passes under the Mesa de Maya, Fig. 141. Again, the continuity of the monocline of the foothills on the western border of the Great Plains, formed upon the upturned edge of the Plains strata, is broken by a number of small offsets, a series of three occurring at Greeley, Colorado, Fig. 117.1 Over the larger portion of the Great Plains the westward-dipping strata end in long, low, ragged, eastward-facing escarpments; the topog- raphy thus has a “‘stair step”? composition, the escarpments stand out as conspicuous risers and the broad, flat, inter-escarpment areas rep- resent the treads. Of such origin are the Flint Hills, the Dakota Sandstone Hills, and the gypsum hills of Kansas and Nebraska; and similar examples occur in many other places. These features are most prominently developed where the hard strata (sandstones) are thick and the soft strata (shales) thin... Where the reverse is true the escarp- ments may be so low as scarcely to be identifiable and the inter-escarp- ment tracts dissected by a maze of branching systemless streams. For example, the Pierre shale occupies many thousand square miles of country adjoining the Black Hills and gives rise to a monotonous landscape of rounded hills thickly covered with grass instead of the more common tabular relief bordered by escarpments.2 The Great Plains of Texas are developed largely upon an extensive limestone stratum (Edwards) which is completely surrounded by escarpments of erosion. In the Llano Estacado, which is the southern end of the High Plains subdivision, the surface is for the most part capped by alluvium from the mountains.? 1 N. H. Darton, Geology and Underground Water Resources, Central Great Plains, Prof. Paper U. S. Geol. Surv. No. 32, 1905, pp. 74-75. 2 Idem, pp. 22-23. 3 R. T. Hill, Physical Geography of the Texas Region, Folio U. S. Geol. Surv. No. 3, 1900, pp. 6-7. GREAT PLAINS 409 STREAM TYPES Two types of streams cross the Great Plains, (1) streams whose head- waters are in the mountains and (2) streams whose headwaters are on the plains. Those streams whose headwaters are in the mountains are supplied with water more or less regularly either through rain or snow or both. In addition, their headwaters often drain glacial lakes and these tend to regulate the flow. The result of the somewhat regular water supply is that although the streams may become very low and even dry up occasionally, they are for the most part through-flowing the entire year. This result appears the more striking when it is known that their plains tributaries are intermittent and feeble and that the master streams suffer heavy losses by evaporation in crossing the semi- arid plains country. A stream of this type is the Arkansas River, whose headwaters are in the Sangre de Cristo, Culebra, and Sawatch mountains, in each of which there are summits exceeding 14,000 feet in altitude. The pre- cipitation along the crests of these high ranges is mainly in the form of snow and amounts to 20 or 30 inches of rain each year. From the foothills of the mountains to Arkansas City the rainfall ranges from 12 to; 35.1nches; it Is 25 to, 35 ..1mches in the last roo miles below Hutchison. Natural storage in the Arkansas basin is limited to a few mountain lakes of glacial origin. The streams are subject to two floods per year — the annual spring floods caused by the melting of the mountain snows, and the summer floods due to cloud-bursts in the foot- hills and plains regions." The Missouri River resembles the Arkansas in that its upper tribu- taries drain a forested region, while the main stream flows through a country almost wholly devoid of forests. The precipitation in the mountainous portion of the basin is mainly in the form of snow, but a great part of the area lies within the arid and the semi-arid regions, and it is probable that the mean annual precipitation throughout the entire basin is less than 20 inches. The river notably decreases in volume by evaporation in crossing the dry plains, though it never disappears, as is the case with many neighboring streams whose headwaters do not reach into the mountains.” In contrast to the Arkansas and the Missouri is the Red River, whose sources are on the plains of northern Texas. The flow of this river is 1 Freeman, Lamb, and Bolster, Surface Water Supply of the United States, 1907-08, pt. 7, Lower Mississippi Basin, Water-Supply Paper U. S. Geol. Surv. No. 247, 1910, pp. 20-31. 2 Follansbee and Stewart, Surface Water Supply of the United States, 1907-08, pt. 4, Missouri River Basin, Water-Supply Paper U. S. Geol. Surv. No. 246, 1910, p. 30. 4IO FOREST PHYSIOGRAPHY very uncertain, the run-off consisting chiefly of flood water from heavy rains. The flow ceases entirely in the late summer and fall of ordinary dry years. The drainage area consists of semi-arid plains varied by small areas of sand hills. If the summer is dry, the flow of the river ceases altogether, although water sufficient for stock always remains standing in pools. During long-continued or unusually heavy rains, or directly after such rains, the river becomes wide and deep, its bottom lands are flooded, and considerable damage is often done to livestock, railroads, and plantations." Those streams that descend the eastern slopes of the uplands south of the Canadian River and near the New Mexico-Texas boundary line are of an extreme type and die out just eastward of the boundary line. Their waters are absorbed in the surface material or gathered in the | form of temporary lakes. It is not until the central portion of the Pan- handle region of Texas is reached that the drainage is through-flowing and reaches the Gulf of Mexico by way of the Red River and other streams. REGIONAL ILLUSTRATIONS NORTHERN GREAT PLAINS Many features of the northern part of the Great Plains in the United States owe their origin and character either to base-leveling or to glacial accumulation or both.?, Westward from the valley of Red River of the North and the prairies of the Lake Winnipeg region is a plains country which rises gradually northwestward to the foot of the Rockies. The chief ascent of this plain west of the basin of Lakes Manitoba, Winnipeg, and Winnipegosis is over an abrupt escarpment developed upon Creta- ceous strata that form the eastern border of the continuation of the Great Plains in Canada. The escarpment is from 200 to tooo feet high, and its sections are named in order from south to north, Coteau des Prairies, the Pembina, Riding and Duck mountains and the Pasquia Hills or Mountains. West of these hills and so-called mountains the broad expanse of plains is broken here and there by valleys and irregularly dissected tracts, but in general the surface appears to be very moderately rolling, rising 4 or 5 feet to the mile to a height of 4000 feet at the foot of the Rocky Mountains in Montana and Alberta. At the beginning of Tertiary time this region became a land surface 1 Follansbee and Stewart, Surface Water Supply of the United States, 1907-08, pt. 4, Missouri River Basin, Water-Supply Paper U. S. Geol. Surv. No. 246, 1910, p. 85. 2 W. Upham, Tertiary and Early Quaternary Base-leveling in Minnesota, Manitoba, and Northwestward, (Abstract) Geol. Soc. Am., vol. 6, pp. 17-20. See also the American Geolo- gist, vol. 14, 1894, pp. 235-246. GREAT PLAINS AII and has remained a land surface down to the present. During this long period (Tertiary) the great tract of country was base-leveled except for isolated residuals, here and there, consisting of remnants of horizontal Cretaceous strata elsewhere eroded. Turtle Mountain on the northern edge of North Dakota and the southern edge of Manitoba is an illus- tration of such a residual. It is about 40 miles long, about two-thirds as wide, and has an altitude of 200 to 800 feet above the surrounding country. It consists of nearly horizontally bedded shales and lignites capped by 50 to 75 feet of glacial drift." West of Turtle Mountain the depth of the Tertiary base-leveling was greater and attained a tremen- dous value on the plains of Montana; in the Highwood and the Crazy Mountains districts the country was planed down 3000 to 5000 feet. These mountains owe their superior elevation to the great resistance of their rocks as compared with the strata wrapping about them; they now stand as embossed forms rising conspicuously above the general expanse of the monotonous plains.? The greater denudation of the western por- tion of the Great Plains was due to greater initial uplift; it appears to have been raised 1000 to 5000 feet. The Tertiary topographic cycle of erosion was closed by a great uplift, after which vigorous stream erosion set in. In the eastern part of the Great Plains region erosion was carried to the point of partial base-leveling and the development of wide flat valleys. The Red River developed a broad plain more than 200 miles in extent from south to north and from 200 to 500 feet below the general level of the country. In Manitoba the Cretaceous beds were eroded down to the underlying Archean and Paleozoic rocks over a large area and the eastern marginal escarpment formed. The depth of post-Tertiary erosion is indicated by the White Hills west of Lakes Winnipegosis and Manitoba, a long line of escarpments. During the period in which this escarpment was forming in Manitoba and in which its tributaries were deeply incised the plains of Montana were partly dissected by the deeply intrenched streams. These plains now present one of the best illustrations of an uplifted base-leveled surface to be found in the country. The surface bevels the strata of the region regardless of changing dip and hardness, con- sequently neither the structure of the rock nor alluviation can be appealed to in explanation of the general topographic uniformity. The 1 W. Upham, Tertiary and Early Quaternary Base-leveling in Minnesota, Manitoba, and Northwestward, (Abstract) Geol. Soc. Am., vol. 6, pp. 17-20. See also the American Geolo- gist, vol. 14, 1894, pp. 235-246. 2 W. M. Davis, The United States in Mill’s International Geography, 1900, p. 756. 412 FOREST PHYSIOGRAPHY interstream spaces are rather flat and give no indication of the marked depth and steep walls of the intrenched streams which are practically invisible until one is almost at the canyon brink. The narrow valley of the Missouri is from 300 to 600 feet below the general level of the plains, and that of the Yellowstone and many others have been cut to comparable levels. GLACIAL FEATURES The map, Fig. 148, shows how small a portion of the Great Plains has been glaciated. Besides glacial features similar to those of the northern ClDuek YEY LEU YY GY Baltic ey as, i / StS Shelby 5 We YY Ley LEGEND ) Joa righton' tox Terminal moraine =| y and older drift, \ ~Chouteauw yy aiyivd : 3/0S3NOT Lyz¥ds Keewatin ice sheet R Wij/\ Area covered by “jj, younger drift = __| S Sun ma Sun | = " Area covered hy ead = | B mountain glaciers YH) Augusta = Ws Area covered by =| glacial lakes MAP OF NORTHERN MONTANA SILOWING TILE GLACIATED AREA Scale of Miles 1) 20 30 «40 113 Z ray Es} Mussells hell Fig. 148. — Note the large proglacial lake west of the Highwood Mountains, caused by the glacial damming of the Missouri River. Its discharge across the northern border of the Highwood Mountains formed Shonkin Sag, a temporary outlet. (After Calhoun, U. S. Geol. Surv.) part of the Prairie Plains may be noted the two terminal moraines which cover the country north of the Little Rocky Mountains. The surface is a rolling plain with rounded flat-topped ridges and broad and low intervening hollows. The larger drift is chiefly Laurentian, but the bulk of the material is made up of quartzite drift from the Rocky Mountains, and consists of well-rounded pebbles and small quartzite bowlders of different colors.! The Keewatin ice sheet extends southward into the northern Great Plains as far as the Highwood, Bear Paw, and Little Rocky Mountains. These elevations stopped its southward progress locally, while on the t Weed and Pirsson, Geology of the Little Rocky Mountains, Jour. Geol., vol. 4, 1806, Pp. 402. GREAT PLAINS AI3 plains between it moved somewhat farther south in great lobes, Fig. 148. The Sweetgrass Hills, almost on the international boundary line (long. t11° W.), were completely surrounded by the ice, above which they pro- jected 2000 feet as great nunataks. Terminal moraines therefore com- pletely encircle the hills, while the higher mountains farther south and east are but partly encircled by moraines. The position of these moraines along mountain slopes and on north-south lines makes it pos- sible to establish the fact that the slope of the ice surface was from 30 to 50 feet per mile. A similar basis is afforded by the disposition of the moraines formed on the margins of lobes that extended up many of the valleys of the region, the divides between (150 feet high) remain- ing uncovered — with a similar result. Terminal moraines fringe the border of the drift-covered country in places only. Elsewhere there is no well-defined terminal ridge, only a broad low rise usually too slight to be noticeable. The terminal moraines are monotonously rough and their hollows contain large num- bers of smaller lakes and ponds. In places the moraines are bowldery ridges, in others they are broad belts of till of variable composition. Near the edge of the glaciated tract the drift is thin, though locally it is from 50 to 200 feet thick. The drainage of the glaciated northern portion of the Great Plains is toward the northeast and east, and ice invasion was therefore bound to block the streams either temporarily by the ice or more permanently by moraines or both. The lakes are of two classes. Those of the first class were in a few instances of great extent, and their floors are noted for the scattered bowlders rafted into position by detached ice blocks. The name Great Falls Lake has been given to one such temporary water body, formed by the ponding of the upper Missouri between the south- ward-facing ice sheet and the northward-facing slopes of the Highwood Mountains. Its outlet along the northern mountain flanks at 3900 feet cut a broad channel nearly a mile wide and 500 feet deep, known as Shonkin Sag. It is one of the most striking topographic features in the region, since it is entirely independent of rock structure and crosses the present drainage lines at right angles. After the disappearance of the ice the terminal moraines continued for some time to act as barriers to the drainage, but outlets were soon cut down and the nearly bowlderless clays of the lake floors exposed. Most of the existing lakes of the region occur in the hollows of the morainic belts and are individually of small extent. Glacial features of considerable topographic prominence are also found upon the northwestern border of the Great Plains of the United 414 FOREST PHYSIOGRAPHY States west of the limit of continental glaciation where valley glaciers from the front ranges of the Rockies deployed, building terminal and lateral moraines and supplying material for local alluviation (p. 312). The glaciers extended as individual tongues of ice down_ the main valleys, but 14 of them united in groups along the base of the moun- tains to form piedmont glaciers of great size which extended out over the plains for from 30 to 35 miles. In other cases the ice extended only part way down the valleys, deposited valley moraines, and caused the formation of related features such as lakes, outwash plains, and valley trains. The piedmont glaciers and the continental ice sheet overrode com- mon ground in two localities— the headwaters of the Marias and St. Mary’s rivers— though these ice bodies were not contemporane- ous. The vailey glaciers had retreated before the continental ice sheet reached its greatest extension, since the deposits of the latter ice body overlie those of the former. The terminal moraines of the piedmont glaciers are of great size, some of them attaining heights of 200 feet and breadths of a mile or more.' BADLANDS OF THE BLACK HILLS REGION Among the badland areas of the Great Plains the largest and most important is in South Dakota and Nebraska on the southern, south- western, and southeastern borders of the Black Hills. The portion of the badlands showing the greatest topographic complexity lies near the southeastern border of the Black Hills between the White and Cheyenne rivers and is known as the Big Badlands. It is continuous with a second badlands areas which extends eastward, southward, and westward along the upper White River forming the high Pine Ridge escarpment which extends through western Nebraska and into Wyoming. Outside the limits of the main badlands area are many remnants of the strata that formerly extended over the region. They now occur as mesas or tables which stand at various heights up to three hundred feet or more above the adjacent basin or valleys. Some of them are of large size and those east of the Cheyenne River have been given individual names by the settlers who occupy them, as Sheep Mountain Table, White River Table, Kuba Table, etc. The badlands were at first thought to be quite uninhabitable, the name itself being derived from the French name ““Mauvaises Terres’’ applied by the early hunters and trappers. The phrase is meant to signify a country difficult to cross, chiefly because 1 F. H. H. Calhoun, The Montana Lake of the Keewatin Ice Sheet, Prof. Paper U. S. Geol. Surv. No. 50, 1906. GREAT PLAINS 415 of the rugged surface and the general lack of water. Later explora- tion and development have shown that much the greater portion of the area within the badlands is level and fertile and covered with abundant grass; and that water may be obtained, especially on the higher tables, by sinking shallow wells in the surface mantle of gravel. As a whole the region has considerable agricultural and grazing importance. The chief factors controlling the development of the badlands topog- raphy have been the great extent of slightly consolidated, fine-grained strata lying at a considerable altitude above the sea in a region of low Fig. 149. — Details of Badlands in Brule clay at foot of Scotts Bluff, Nebraska. (U.S. Geol. Surv.) rainfall and sparse vegetation. The scarcity of deep-rooted vegetation enables the soft material to be rather easily eroded. While short grasses are abundant over large areas they do not have deep root penetration and do not form a sufficiently continuous cover to prevent cutting. There is a small amount of vegetation of a higher order, but it is even less effective than the grass in preventing the formation of gullies. A few gnarled cedars occur on the highest points and bushes of various kinds occur in the valley floors in favorable places, but they offer little obstruction to the development of the gulches and canyons which diver- sify the scarped margins of the area. 416 FOREST PHYSIOGRAPHY It is also noteworthy that the rainfall is more or less concentrated into heavy showers of short duration and these act most vigorously in denuding the surface, forming channels, and transporting accumulated rock waste. The clays which compose a large part of the badlands expand greatly when wet and contract when dry (p. 415), so that their surfaces tend constantly to break up and form an easily eroded mass of material. Frost action tends toward the same effect, with the result that the clay is constantly being loosened in the dry as well as in the wet season, and is thus prepared for rapid erosion when rain falls. The topographic complexity of the badlands is explained chiefly by the unequal resistance to erosion on the part of.alternating beds. In- dividual layers vary horizontally as well, and the softer portions are worn back into the form of gulches and alcoves while the harder portions remain projecting as spurs or headlands. A hard layer tends to persist longer than a soft layer above or below it, thus forming typical scarp and terrace topography. In many cases the hard layer is undercut to such an extent as to develop a precipice. Joints are developed here and there in the clays and these tend to accelerate erosion along vertical planes with the result that cave-like excavations are formed at the heads of vertical walled gulches. Many beds contain large numbers of concretional masses whose resistance is greater than that of the surround- ing material, thus tending to the production of a surface whose irregu- larities reflect the irregularities of the concretions. Among the important topographic elements of the badlands are the alluvial fans which cling to the base of every pillar, mound, or table, and have a large total extent. Their surfaces have low gradients and they represent the lag of transportation behind disintegration. Most of the streams of the badlands area are intermittent. They have a continuous flow for a short time after the rainy season but their sources rapidly fail, and soon nothing is left in their channels but bars and banks of sand and silt, and a few trifling pools of water, either so strongly alkaline as to be of little value, or so turbid as to have the consistency of mud. Only two streams have a continuous flow, Cheyenne River and White River; but their flow varies greatly in volume, being high in the rainy season, when their tributaries are full, and very low in the dry season, when their tributaries fail. Many streams, especially the streams of medium size, flow in box-like trenches, the material of their banks standing up in bluff-like form as about the border of the area.! 5; 1 For a summary description of the badlands of South Dakota, illustrated by well-selected photographs, see The Badland Formations of the Black Hills Region, C. C. O’Harra, Bull. S. D. School of Mines No. 9, 1910. GREAT PLAINS ALT HIGH PLAINS The map, Plate IV, indicates the extent of the High Plains, the most important subdivision of the Great Plains province. The interior topo- graphic and drainage features of the High Plains are typically represented in the Llano Estacado (Staked Plains) of Texas, the surface of which appears extremely flat to the eye, though it actually slopes eastward at the rate of 8 or 10 feet to the mile. Only gentle depressions occur “ningdspee INS Guthrie Fo? te, aS * gy STockton\y org! \Presidio Fig. 150. — Physiographic subdivisions of Texas and eastern New Mexico. (Hill, U.S. Geol. Surv.) here and there — the products of underground solution and of unequal wind erosion. Local storm floods tend to level these irregularities by filling the hollows and eroding the surrounding surface.! The High Plains strata were laid down as a series of great compound alluvial fans or a piedmont alluvial plain at the eastern base of the Rocky Mountains. The eroded bedrock floor upon which the materials were deposited and the eroded condition of the materials themselves 1 R. T. Hill, Physical Geography of the Texas Region, Folio U. S. Geol. Surv. No. 3, 1900, p. 6, col. 3. 418 FOREST PHYSIOGRAPHY imply a succession of changes in the character of the stream action from degradation to aggradation and back again to degradation. These changes harmonize with the conception of climatic changes of known ee Se Nereieseiierpeenecemene= === Fig. 151. — Ideal structure of the Tertiary deposits of the High Plains. The dark band indicates the position of a partly consolidated portion of the section known as a mortar bed. (Gould, U. S.-Geol. Surv.) 1 occurrence, the dry Tertiary (aggradation) being succeeded by the wet Pleistocene (degradation), which was in turn succeeded by the dry Fig. 152. — Typical view of the High Plains of western Kansas. (Gilbert, U. S. Geol. Surv.) Present in which the streams are agerading their valley floors.! It is also conceivable that cutting power may have been gained or at least increased (1) by broad uplift on the west which would increase the 1 D. W. Johnson, The High Plains, zoth and 21st Ann. Repts. U. S. Geol. Surv., 1898- 1899, 1899-1900. GREAT PLAINS 419 stream gradients and (2) by a cooling of the climate so as to reduce the evaporation. The greater part of the Tertiary deposits of the High Plains consists of clay, sandstone, and conglomerate, with clay largely predominating. The materials are usually arranged in a heterogeneous manner, as shown in Fig. 151, a characteristic dependent upon their origin as coarse channel or finer flood-plain deposits.'. A large part of the deposits is composed of smooth rounded white or yellow quartz grains locally cemented by lime into coarse rough sandstones, but more commonly unconsolidated and in places blown by the winds into dunes. In many instances the sand is mingled with small pebbles and clay cemented with lime, forming the so-called “mortar beds.”” The pebbles vary greatly in size, shape, and color, and generally occur in more or less lenticular cross-bedded layers, some of which are at least 50 feet thick. In many places they are mixed with fine sand and locally are scattered through the clay formations. The clays are generally white, but locally buff and pink in color. In places the lime cements the clay into irregular and grotesque forms, which, being harder than the surrounding forma- tions, are left standing in relief by erosion and form picturesque elements in the local scenery.? The average thickness of the Tertiary deposits in eastern Colorado and western Kansas is 200 to 300 feet, but increases to nearly 1000 feet in portions of western Nebraska and southwestern Wyoming. Origi- nally the entire region was probably covered with late Cretaceous deposits that extended far up the flanks of the Rocky Mountains, the Bighorn Mountains, and the Black Hills, a conclusion indicated by the occurrence of outliers of these deposits at high altitudes, the interven- ing portions having been extensively stripped off. PHYSICAL DEVELOPMENT After the great Rocky Mountain System on the west had been out- lined and during late Tertiary time there was a long period in which streams of moderate gradient drained across the central portion of the Great Plains. Locally, extensive lakes were formed, but the larger part of the deposits (Oligocene) of that time consists of coarse sand- stones, sands, etc., deposited on river flood plains and aggrading stream channels. Aggradation took place chiefly in Oligocene time, and the area of Oligocene deposits, Plate V, extends across the eastern part of 1C, N. Gould, Geology and Water Resources of the Panhandle, Texas, Water-Supply Paper U. S. Geol. Surv. No. 191, p. 33- 2 Tdem, p. 32. 3 N. H. Darton, Geology and Underground Water Resources of the Central Great Plains, Prof. Paper U. S. Geol. Surv. No. 32, 1905, pp. 169-170. 420 FOREST PHYSIOGRAPHY Colorado and Wyoming, western Nebraska and South Dakota, and probably northward into Canada. The streams of this time shifted their courses from side to side and laid down a sheet of débris in the form of low-grade and confluent alluvial fans of vast extent and with a thickness in places of tooo feet. The period of deposition ceased with the Miocene, and uplift and erosion followed which removed large por- tions of the accumulated material. During this period of deposition (Oligocene) in the northern part of the High Plains the southern part was probably a region of erosion, but with the uplift and erosion of the northern High Plains (at the close of the Miocene) the southern por- tion became a region of deposition and the streams began depositing thin mantles of late Pliocene sands in southern Colorado, southern Nebraska, Kansas, and still more southerly localities.! It is important to recognize that these changing conditions of late Tertiary deposition and erosion first in the north and then in the south were determined by differential uplift. The uplifted region suffered erosion, the depressed region suffered deposition. It is equally important to recognize the fact that these occurrences took place in the Tertiary under more or less stable climatic conditions and before the advent of the glacialice. We have in these conditions clear evidence of broad differential uplifts and depressions which produce alter- nating conditions of aggradation and degradation over large areas during the Tertiary period, and which were, so far as we may judge, wholly unrelated to climatic influence except the general influence of aridity to leeward of the high Rocky Mountain Cordillera. The broad changes of later date are, however, undoubtedly to be associated with climatic change, though the exact locality in which erosion or deposition was strongest was probably determined by crustal movement. During the early Pleistocene uplift of the land there also occurred an increased precipitation which resulted in widespread degradation of the preceding deposits; the Tertiary deposits were entirely removed in the eastern portion of the area and widely and deeply trenched in the western portion.2. There was deep erosion about the Black Hills dome, and the High Plains, whatever their extent in that direction, were largely removed and their northern edge left as at present in the great escarpment of Pine Ridge facing the Black Hills uplift. The streams of Pleistocene time also cut deeply and removed widely the high plains of Nebraska, Colorado, Kansas, and Texas, though wide areas of tabular surfaces are still exposed .3 Descriptions of various portions of the High Plains vary greatly with reference to the present character of the stream work. It is fre- quently asserted that the High Plains are a region of denudation, although as frequently one finds the rivers described as building up the bottoms of their valleys. This contradiction is only apparent. The greater portion of the High Plains and many portions of the Great Plains are being denuded by the ramifying headwater streams, and the 1 N. H. Darton, Geology and Underground Water Resources of the Central Great Plains, Prof. Paper U. S. Geol. Surv. No. 32, 1905, pp. 185-186. 2 W. D. Johnson, The High Plains and their Utilization, 21st and 22nd Ann. Rept. U. S. Geol. Surv., pt. 4, 1899-1900, t9g00-1901. 3 N. H. Darton, Geology and Underground Water Resources of the Central Great Plains, Prof. Paper U. S. Geol. Surv. No. 32, 1905, pp. 186-188. GREAT PLAINS A421 detritus is being carried down to the valley bottoms, where it accu- mulates in the form of broad valley flats or flood plains which to a large extent control the curves of the rivers, as along the valley of the Platte. Such aggradation is a normal result of the return to the drier conditions prevailing both now and in the Tertiary. The process will continue until a gradient is established which will allow the streams to carry all their waste to the sea. Dissection by the streams of Pleistocene time caused such a degree of valley cutting that the present aggrada- tion on the floors of the valleys is small compared with the degradation that preceded it. Consequently the tributaries of all the master streams of the High Plains region join valleys far below the general level. The tributary streams are therefore eroding the surface actively, and they drain by far the larger portion of the High Plains. Active erosion has stopped completely and active aggradation taken its place on almost all the larger valley floors. BORDER TOPOGRAPHY In some localities the erosion of the border of the High Plains has given rise to the formation of “holes” with local badland topography. A typical instance is Goshen Hole on the border of Wyoming and Nebraska. The formations which outcrop in this area consist largely of clays and sandstones. The clay (Brule) is easily eroded in such manner as to keep the sandstone (Arikaree) covering it in constant retreat as a nearly vertical wall about the edges of the Hole, a process greatly hastened by the issuance of ground water at the line of contact of the two formations. The escarpment separates the dissected low- land of the floor of the hole from the upland or general undissected surface of the High Plains. It is best defined upon the western side of Goshen Hole, owing to the absence of streams upon this portion, though the escarpment is here in constant retreat as the result of the sapping action of the ground water at the heads of the minor streams to which it gives rise. Upon the lowland constituting the floor of the hole are small mesas and tablelands, remnants of the upland that once occupied the region and that has since been worn away. Wherever the relations of drainage and geologic formations are such as not to bring the streams’ below the soft strata or where the upper strata are soft and the lower strata hard, either a lowland has not been formed or, if formed, its bor- ders are not so well defined as in those cases where the softer members underlie the harder and the drainage has been developed distinctly below the level of the soft formations. The eastern border of the High Plains in Texas and Oklahoma is in 422 FOREST PHYSIOGRAPHY most places a distinct scarp 200 to 500 feet above the eroded plains on the east, though ordinarily the descent is more gradual and takes place in a belt 5 or 6 miles wide —a belt with distinct topographic features in high contrast to the flatter plains to east and west.' This escarp- ee Fig. 153. — Jail Rock with the valley of North Platte in the distance, looking east. The capping stratum is sandstone (Gering) while the slopes are of clay (Brule). Typical border topography of a portion of the High Plains. (Darton, U.S. Geol. Surv.) ment is locally known as “The Breaks,” Fig. 155. It occurs along the valley margins, passing in broad eastward-looping curves from one drainage system to another, and is characterized by badland erosion Fiz. 154. — Eastern erosion escarpment of the High Plains of Texas, Oklahoma, and Kansas. The eroded bed-rock surface has been covered with Tertiary deposits and later reéxposed in part. (Johnson, U.S. Geol. Surv.) forms, short ridges, steep talus slopes, isolated buttes and peaks, and an intricate system of narrow V-shaped valleys that sometimes develop into impassable canyons, Fig. 154. It is a region very difficult to trav- 1 C. N. Gould, Geology and Water Resources of the Eastern Portion of the Panhandle of ~ Texas, Water-Supply Paper U. S. Geol. Surv. No. 154, 1906, p. 9. GREAT PLAINS 423 erse, and can be crossed with a wagon only at infrequent intervals and by specially selected routes. The marginal escarpment is most typical along the Canadian River and in Palo Dura Canyon in Armstrong County, Texas. ' Fig. 155. — Details of form, eastern border of the High Plains, Texas. The scarp is known as “ The Breaks of the Plains.” (Hill, U. S. Geol. Surv.) The eroded plains east of “The Breaks”’ have suffered such extensive dissection that the Tertiary and Pleistocene deposits have been en- tirely removed from their surface and the streams are actively dissect- ing the underlying Red Beds. The region is a rolling plain into which the streams are cutting rather deep steep-sided valleys, and outlining hills generally of conical shape but ‘often elongated and attaining heights of 200 to 800 feet or more. These hills are capped by resistant ledges of sandstone, gypsum, or dolomite that have resisted the erosion which swept away the softer clays and shales above them.! SAND HILLS AND LAKES The sand hills of the Panhandle region of Texas range in size from small mounds to ridges 30 or 4o feet high. They are oval, crescentic, or elongated in shape and extend in various directions. They are inter- spersed with broad, shallow, basin-like depressions from 1 to ro acres in extent, probably representing great “blow-outs” or wind-eroded hollows. In a few localities there are migratory dunes, as on the west 1 C. N. Gould, Geology and Water Resources of the Eastern Portion of the Panhandle of Texas, Water-Supply Paper U. S. Geol. Surv. No. 154, 1906, pp. 9-10. 424 FOREST PHYSIOGRAPHY side of the Canadian River in Roberts County. The dune sands are derived from sandstone ridges chiefly, but also in part from river sand. The rainfall on an important part of the southern High Plains is collected into shallow saucer-shaped depressions known as “lakes” or playa lakes, scattered irregularly about. The depressions vary in size from a diameter of a few feet and a depth of a foot or two to lakes several hundred rods in diameter and from 20 to 4o feet below the general level. Some of these depressions are more than a square mile in area. Many lakes in the depressions are perennial and afford an Ne Shute Fe aaa [) Ofvoudens cp ce | oe J ax, A nile % Ho. ri : NV Fig. 156. — Precipitation in the Texas Region. I, over 50 inches; II, over 45; III, over 40; IV, over 35; V, over 30; VI, over 25; VII, over 20; VIII, over 15; IX, over 1o. (Hill, U. S. Geol. Surv.) abundant supply of water for stock. Others are ephemeral, that is, they are filled during a period of rain, but soon dry up; and still others con- tain water only at long intervals. Before wells were constructed and in the early settlement of the country the lakes were often in intimate relation to the location and welfare of settlers, for they constituted | practically the only water supply. Wagon trains in crossing the Llano Estacado camped beside them, and they were centers to which cattle were driven until comparatively recent times.1 1 C. N. Gould, Geology and Water Resources of the Panhandle, Texas, Water-Supply Paper U.S. Geol. Surv. No. 191, 1907, p. 50. GREAT PLAINS 425 The sand-hill country is not limited to the High Plains but is found (1) in many cases with limited development along the valley floors where sands deposited in the stream channels during high water become available to the winds during the low-water stages, as along the Arkansas, and (2) over the outcrop of unconsolidated material. The largest tract lies in Nebraska! and was developed upon the outcrop of a sandy uncon- solidated formation which the present sand cover largely conceals. The thickness of the sand cover is rarely over too feet and generally much less. The total area of the tract is about 18,000 square miles. A peculiar feature of this great area is the relative absence of streams, except such as are supplied from outside sources. In spite of the small surface drainage there is considerable spring flow and seepage along the main river valleys. Most of the rainfall is absorbed by the loose dry sands and percolates downward, to be added to the ground water, which is surprisingly large in amount and is in general of good quality. The large amount of ground water is reflected in the numerous lakes scattered through the sand-hill country. Since the lake surfaces represent the surface of the ground water the lake levels rise and fall in response to the variable rainfall just as does the ground water. In times of light rainfall they are smaller and shallower; in times of beavy rainfall they are deeper and broader. Some of them disappear in extremely dry seasons; others are so permanent that their waters are stocked with fish. They are of great importance as a source of stock water, since the bunch- grass of the sand hills gives the area high value in respect of grazing, the chief industry of the region VEGETATION Bunch-grass is the characteristic vegetal growth of the High Plains. Its growth is extended by roots and more rarely by seeds. The dryness and coarseness of the surface soil and its wind-swept condition render reproduction by seed very difficult. The result is not a turf so dense that the ground is not visible but a series of tufts or bunches separated by large and small bare spaces.” It is the general conclusion of students of the plains streams that floods are growing in volume and frequency, a state of things that is thought to be related largely to the breaking up of the protective grass cover by over-pasturage in the headwater regions. A grass cover has almost as great influence in checking run-off and favoring absorption 1G. E. Condra, Geography of Nebraska, 1906, pp. 85-04. * Report upon Geographical and Geological Explorations and Surveys West of the rooth Meridian (Wheeler Surveys), Geology, vol. 3, 1875, p. 606. 426 FOREST PHYSIOGRAPHY by the soil as a timber cover, and this relation is of such importance throughout the Great Plains region as to make it a point of great interest to the forester interested in general conservation, for the com- bination of grass-land and timber is thoroughly practical. — - In the middle courses of the Great Plains streams and on numerous tributaries in the mountains there is a protective timber covering. In a similar way a timber growth is found on the borders of the escarp- ments. and its maintenance is a matter of the liveliest concern to every | OK PASEO i Brownsyil Fig. 157. — Vegetation of the Texas regions. 1, Atlantic forest belt; 2, Rocky Mountain forest; 3, Chap- arral; 4, Black Prairie; 5, bolson desert flora; 6a, Grand Prairie; 6b. Great Plains; 7, transitional, with plains, prairie, and Atlantic flora; 8, coast prairies; XXX, yucca belts. (Hill, U. S. Geol. Surv.) one in any way related to the régime of the streams. The amount of washed soil running off in the floods that follow cloud-bursts in the Great Plains is enormous, especially where trees are absent and the grass covering thin; the maximum extent of grass and forest cover ought therefore to be maintained. This is seen especially in the lignitic belt of Texas, which is a rough broken country with sandy soils. But for the shortleaf and post-oak forests which cover them the soils would be washed in quantities from the steeper slopes. Annual plants with a superficial root system can not flourish in such soils without abundant rainfall. More than that, soils in these positions have little capacity GREAT PLAINS 427 for retaining moisture, and forest growths are slow in beginning. The hills bordering the Edwards Plateau would become arid and worthless if they were stripped of their forest cover. The inch-deep soil débris would be rapidly washed away and the restoration of the forest become impossible. The Great Plains contain but little timber, though the supply is in many cases sufficient for local and limited use. In the Camp Clarke district of Nebraska the ridge extending west from Redington Gap bears scattered pine trees of moderate size, and there are also a few Rocky Mountain pines (Pinus ponderosa) on the slopes ascending to the high table at the southern margin of the district. The largest are from 1 to 2 feet in diameter. A moderate number of young pines begin growth in favorable situations on the ridges, though few of them attain maturity. The zone of cottonwoods is characteristic of most western streams, though it is absent along North Platte River, where only a few small trees and bushes are found. The principal deciduous growths are found in ravines; they include cottonwood, box elder, wild plum, and a few other varieties.! The Great Plains include much of the central treeless region of the United States where forest planting has been a part of the progress in agriculture, so that the areas of most extensive agricultural develop- ment are also those where greatest tree planting has taken place. Ap- proximately 840,c00 acres have been planted in the central treeless region, but the best conditions can be obtained only by planting about 14,000,000 acres in all. Although forest planting had been in a period of decline in this region, it has recently revived. It has been demon- strated that about 5% of the prairie region should be forest covered, and farther west, in the Great Plains proper, 3% could profitably be devoted to tree growing. South Dakota has planted about 122,000 acres; Nebraska, about 192,000 acres; Kansas, 175,000 acres; Oklahoma, 21,000 acres; Texas,.13,000 acres. The chief purpose of such tree planting has been for shelter, for post production, for protection from the hot winds of summer and the blizzards of winter, and to a limited extent for fuel.? A point of great interest in connection with the limited tree growth of the central plains is the possibility of an earlier and more extensive timber cover which may be restored by proper effort. If the rainfall was more abundant for a time during and after the retreat of the ice, it follows that timber may have grown upon areas now too dry to support it. The 1 N. H. Darton, Camp Clarke Folio U. S. Geol. Surv. No. 87, 1903, p. I. 2 Yearbook Dept. Agri., 1909, pp. 340-342. 428 FOREST PHYSIOGRAPHY thought is the more plausible when we recognize the great length of time involved in the disappearance of the ice sheet, long enough, we may suppose, for considerable progress to have been made in natural reforestation. The conclusion appears to be supported by the finding of remnants of old pine forests in geologically recent deposits in the valley of the Niobrara, 50 miles southeast of existing forests of pine. It has been concluded that these forests formerly extended down near the mouth of the Niobrara.!. Likewise in the sand-hill country of Nebraska the stumps of cedars and pines— trees now confined almost wholly to the river bluffs —are found some distance away from the rivers, thus suggesting former more extensive tree growth.” The fundamental condition governing the treelessness of the Plains and the Prairies is the deficiency of rainfall in spite of the fact that the larger portion of the rainfall of the central plains region occurs in the summer or growing months, Fig. 22. Beyond the Mississippi the rainfall shades off rapidly; in sympathy with this increasing dryness is the gradual thinning out of the forest and its final disappearance. In the transition belt it is obvious that secondary forces will hold the balance of power. Among these undoubtedly the most important are the fine- ness of the soil, forest fires started by the Indians and by lightning, the prolonged droughts of exceptionally dry years, the high rate of evapo- ration of soil water, and the low humidity. The importance of the fineness of the soil has been strongly urged,’ since almost all forms of vegetation require an aerated soil, and this is true especially of all but a small number of forest trees. Prairie soils are notably fine and compact, especially in their natural state, and this tends to keep rainfall near the surface for a longer time than in the case of coarse soils, hence a larger part of the rainfall is reévaporated. The deeper-lying tree roots are thus deprived of a large part of the rainfall, besides being deprived by grass roots of such rainfall as is left available to plants. Prairie and forest fires are of frequent occurrence, and before the coming of the whites were often started by Indians either in sheer ignorance, carelessness, and wantonness, or for purposes of war and the chase. The number of thunderstorms that develop on the plains dur- ing the summer months is excessive. Lightning starts a fire in the dry half-withered grass and this spreads with great rapidity and is often not extinguished by the ensuing rain, which may be very light. Indeed 1S. Aughey, U.S. Geol. and Geog. Surv. of Col. and Adj. Terr. (Hayden Surveys), 1874, p. 266. 2 G. E. Condra, Geography of Nebraska, 1906, p. 93. 3 J. D. Whitney, Plain, Prairie, and Forest, Am. Nat., vol. 10, 1876, pp. 577-588 and 656- 667. GREAT PLAINS 429 many summer thunderstorms are not accompanied by rain and fires burn unchecked. That this was formerly an important natural agent is shown by the extension of the natural forest upon prairie land in the Edwards Plateau of Texas.! Settlement has stopped the periodic burn- ing of the grasses, which were harmed by fires much less than were the shrubs, the vanguard of the timber vegetation. The over-pasturing of the ranges of this region has prevented the grass from forming a con- tinuous sod, thus throwing further influence in the direction of the timber growth. In recent years the results have been marked. The shrubs have come in rapidly, trees have grown in their wake, and there is now in progress a gradual but extensive transition from grass to woody growth. The prolonged droughts of exceptionally dry years are effective in preventing the germination of the seeds of forest trees and in harming tender seedlings. At least three-fourths of the rain that sinks into the ground never gets beyond the surface layer of two feet,? while the increas- ing distance to the ground water with increasing aridity is well known. When these naturally hard conditions are enforced in exceptionally dry years the roots of seedlings and mature trees are confronted with grave difficulties in their search for a water supply. The water in the surface layer is soon evaporated in large part or left in the form of a film sur- rounding the soil grains where it ceases to be supplied by capillarity and gravitational flow to the root zone of forest trees. Evaporation is hastened by the high summer temperatures of the plains and the con- tinued sunshine and the rather steady and often high winds. It should not be forgotten in the analysis of the problem that the under- lying condition is regional dryness. Were the rainfall heavy, or even moderately heavy, a forest cover would undoubtedly exist in the plains country. Besides this factor the others appear secondary, yet the sum total of the secondary forces definitely restricts the forest. (1) There are far windier situations on forested mountain slopes that one may find on the plains, (2) forest fires do not eliminate trees from many relatively dry regions, (3) heavy clay lands bear luxuriant forests where rainfall and temperature conditions are normal. But when the condi- tions of rainfall are critical the balance of power is held by otherwise feeble influences. That the cause is not wholly climatic is reasonably inferred from the success with which tree planting has been conducted in the treeless 1 W. L. Bray, The Timber of the Edwards Plateau of Texas, Bull. Bur. For. No. 40, 10904, PP. 14-15. 2 F. H. King, Productivity of Soils, Science, n. s., vol. 33, 1911, p. 616. 430 FOREST PHYSIOGRAPHY settled portions of the western country where windbreaks, snowbreaks, thrifty groves in city parks, etc., testify to the possibilities of at least a restricted reforestation. Even the sand dunes of Nebraska appear to be reclaimable, as shown by the at least temporary success that has attended the reforestation efforts of the government. The whole plains country can not be reforested since the greater part of it is too dry, but a large part will support a limited forest cover and the Prairie Plains can be made to grow trees almost everywhere, where now the growth is limited, at least toward the western border, to the valley floors and borders, and to tracts of porous soil with a good water supply. CHAPTER XXIII MINOR PLATEAUS, MOUNTAIN GROUPS, AND RANGES OF THE PLAINS COUNTRY EDWARDS PLATEAU THE combined Edwards Plateau and Llano Estacado (the portion of the High Plains that lies in western Texas) is the most extensive relief feature of the non-mountainous part of Texas, with an area of about Fig. 158. — Llano Estacado, Edwards Plateau, and adjacent territory. The central denuded region was once covered with the Edwards limestone now exposed about the border as a frayed escarpment of erosion. (Hill, U. S. Geol. Surv.) 431 432 | FOREST PHYSIOGRAPHY 60,000 square miles. The two merge into each other and no sharp line can be drawn between them. They are surrounded on all sides by pronounced escarpments. The eastern margin is an escarpment of headwater erosion, the southern margin is the dissected Balcones fault scarp, and the western and northern margins descend steeply to the drainage grooves of the Pecos and the Canadian respectively. While the Edwards Plateau continues the High Plains of western Texas or the Llano Estacado southward, it presents structural conditions wholly different from those found in the High Plains and, except for the general plain-like character of much of the surface, is quite different in its topographic character. It should be regarded as a subdivision of the Fig. 159. —Summits of the Lampasas Plain, Texas. (Hill, U. S. Geol. Surv.) Great Plains province and not as a portion of the High Plains. The surface of the Edwards Plateau is capped by the Edwards limestone, the most conspicuous and extensive sedimentary formation in Texas (and a large part of Mexico) inland from the coastal plain. The Edwards limestone is also important topographically, for its harder strata resist erosion to a greater degree than associated formations, and it is on this account the chief structural factor in the formation of the scarps, mesas, and buttes of the Grand Prairie, Edwards Plateau, and large portions of the limestone mountains of Mexico.! The wide, flat, upper surface of the Edwards Plateau is terminated on the borders of the province by a pronounced descent, a ragged scarp due to the dissection of the draining streams whose irregular interlock- ing headwaters have cut up the border into innumerable circular mesas 1 Hill and Vaughan, Nueces Folio U. S. Geol. Surv. No. 424, 1808, p. 3. PLATEAUS AND RANGES OF THE PLAINS COUNTRY 433 and buttes. The descent from the level plateau to the canyons is over a cornice layer of hard rock that weathers into a nearly horizontal ver- tical bluff. At intermediate levels on the walls of canyons and valleys cliff makers occur; the intervening spaces between cliff makers consist of more easily weathered beds that yield a sloping talus. Fig. 155 represents typical features in the eastern border of the plateau. The drainage of the plateau summit except in the case of the largest streams is of the intermittent variety, and in places even some of the /s fA I\é EGS Ly if = = Fig. 160. — Summits of the Callahan Divide on the Great Plains of Texas (Central Province of Fig. 158). (U. S. Geol. Surv.) evaporation, which is especially strong under the cloudless skies of western Texas, but in larger degree to absorption in the waste-filled valley floors or in the fissured limestone. The smaller streams have been called streams of gravel rather than streams of water, so clogged do their valley floors appear. The streams draining the border of the plateau have a very much more regular flow than might be expected because of the large number of springs which are fed by fissures and underground channels into which the absorbed water is directed. A subdivision of the Edwards Plateau is the Stockton Plateau west of the Pecos and east of the front ranges of the Trans-Pecos region. The southern border of the Stockton Plateau is tilted and faulted into low monoclinal blocks, while the scarps of the northern border overlook the Toyah basin. Formerly the Edwards Plateau extended farther eastward, a condition A34 FOREST PHYSIOGRAPHY shown by the erosion remnants now occurring as outliers beyond the eastern border of the plateau. These remnants consist of circular flat- topped hills and groups of hills on the interfluves. The summits of the outliers are all of the same geologic formation as the adjacent plateau and are in vertical alignment with the normal coastward slope.’ The altitude of these outlying mesas is about 500 feet above the principal stream courses and about 250 feet above the surrounding plains. They form less than 10% of the total area, although they are widely dis- tributed. The principal group occurs on the 31st parallel on the divide between the Brazos and the Colorado and is known under the collective name of the Callahan Divide. As a group these remnants represent a former broad topographic level that once extended from the mountains of the west to the eastern border of the Grand Prairie of Texas, p. 493. During and since the Tertiary this old level was largely destroyed in the central region of Texas and two opposing escarpments formed — the eastern border of the Edwards Plateau and the western border of the Grand Prairie — which are retreating from each other at the present time. Although the Balcones escarpment is commonly referred to as a fault scarp it should be noted that a portion of the descent is to be attributed to the increased steepness of dip along the front of a mono- clinal fold which has been still further accentuated by erosion, Fig. 162. The northeastern extension of the Edwards Plateau forms a third sub- division of the province. Its surface is structural in origin, as indicated by the large number of flat-topped remnantal summits which dominate the tract. They are developed most extensively on the divides between the drainage lines and in places have sufficient height to be called mountains. In general the plateau remnants are bordered by bare white limestone cliffs above and by gentler waste slopes below. The flat summits consist of weathered limestone in places without soil, in other places bearing a thin but rich soil cover. The largest continuous area of these remnants is called the Lampasas Plain which extends for 115 miles from the Colorado in Travis County to the northeast corner of Comanche County, Fig. 160. As shown in Fig. 159, there is an intimate relation between the topography and the geology on the one hand and the vegetation on the other. From the scrub oak and post oak growths of the summit with its thin soils one passes in downward succession over the nearly soilless cliffs with a growth of shin oak to the deeper soils of the sandy formations with their good forests of black jack and post oak.? 1 R. T. Hill, The Physical Geography of the Texas Region, Folio U.S. Geol. Surv. No. 3, 1900, p. 6, col. 4. 2 R. T. Hill, Geography and Geology of the Black and Grand Prairies, Texas, 21st Ann. Rept. U.S. Geol. Surv., p. 7, 1899-1900, p. 77 et seq. 435 PLATEAUS AND RANGES OF THE PLAINS COUNTRY “og ‘ULysne, pur TH) “Isl outeszx9 uO FeYyo UNSNY ya] AuUIAI}X9 UO dUOJSOUT] SpIVMpY “Yout a4} pieMoy} dip jo aSPoIDUl pur ou0z ynvy jo s[tejop d}ON Wys aWaI}xXI UO IMIVI Wr ‘ O yaa. “d sv Y}MOIS }SoIOJ oUIES ‘sIOQWITT, SSOI) UO}SOM ‘spurs AWUITY, ‘23 ‘yodiunt jo yyMOIZ YIM sadojs 9u0}SoUNT] WOLIV Cr ‘Sulmvaq-soyVM ‘purl }MIy pood puv uojj09 lv} pur ulojsey ‘pues Axnjeg ‘d ayouermog ‘7a *yxeo ysod pur ‘ ay} 0} 19}eM puNoIsIapuN pue sd * you pau IPM au eseneunses estates 3 7) uipjnog (ang *J095) "SA ‘TIHH) “d SB YpAMoIF ysaroy aures ‘sToquL], SSO) ‘soonjd AUvU UT PUR] 9[IGIF “YSU sadIvrg INU]? M ‘UOT}LULIOJ NUE MA “J yo daly “Yvo qnsos jo YIMoIS ‘svare as1v] ul }dadxe ssay]lOs fulv[g wD (CAINS “JO9) dy} 0} Jaaj COOS ‘savas [VoT}IOA puv [equozoy{ *(qysI1) ysvo Joya. UL auoZ yNey souosfeg “Wo, uo nesywid spivapy — “Zr ee am as as yaasgQaeg v3 uraysa\y {spurs ApUIAy, 4 “[Los poos ‘sjods otread qjeurs “71 *ZuLIvdq-Io}VA {SIIAV] AfIeut pappoediozul WOIF S[LOS aftjzoy [eUOISLIDO sOvUINS PUL ‘yvo dAl] ‘youl yovjq puv xvo ysod YIM payzsosoy ‘sraquit], SSO1) U1d}S9M "yvo UTYS JO Y]MOIS ‘sadojs Ssaq[los ‘Aypeyo sauoysoul] Weed svseduiey] Jo aplarp yuruns ‘ouojsouny vursded ‘a *Z0]0a3 ‘Kqder30do} ‘Arjsarxoy ‘OINyNI[IBV JO WOL}L[eI SuLmoys ‘ure[g yD sesedurvy oy} fo aplarp B Jo uoryeyuasaidar orywuuviserq — “19t * B11 BLT 436 FOREST PHYSIOGRAPHY PHYSIOGRAPHIC DEVELOPMENT The main mass of the Edwards Plateau was elevated into permanent dry land at the close of the Cretaceous period. During the Eocene period there was extensive erosion and the upper Cretaceous forma- tions (3000 feet) were swept away. Near the close of the Eocene, fold- ing and faulting took place and the Balcones fault scarp that limits the plateau on the south was outlined. During the Miocene. the plateau was still further stripped and the Edwards limestone that now forms the surface of the plateau was exposed. Active erosion has continued from the early Pleistocene down to the present day. At the time of its ap- pearance above sea level at the close of the Cretaceous the Edwards Plateau region probably had topographic features similar to those ex- hibited to-day. The strong differences between the alternating strata resulted in wide stripping wherever the upper surface of a harder stratum of wide extent was exposed. SOIL COVER The soils of the Edwards Plateau are residual and consist in large part of chert nodules that have resisted solution and ordinary wear much better than the soluble limestone. With these are found impuri- ties common to limestone everywhere and even a certain proportion of the calcareous element. Upon the higher levels where flattish summits prevail the soils are good, but along the stream courses they are usually rough and stony. VEGETATION The most important topographic elements of the Edwards Plateau from the standpoint of vegetation are (i) the flat-topped summits of the plateau on the divides, (2) the “breaks” or scarps and related slopes of its ragged canyoned border, known locally as “the mountains,” and (3) the streamways and their tributaries. The ragged escarpment which borders the plateau rises from 400 to 1000 feet above the coastal plain, and it is chiefly on this border that rainfall occurs.! The Edwards limestone has a high absorptive capacity pecause of its low dip and the extensive system of fissures and caverns developed in it. These structures operate also to convey a large part of the water to the deeper strata, from which it discharges as spring water on the valley 1 W.L. Bray, The Timber of the Edwards Plateau of Texas: its Relation to Climate, Water Supply, and Soil, Bull. U. S. Bur. For. No. 40, 1904, p. 9. PLATEAUS AND RANGES OF THE PLAINS COUNTRY 437 margins. The border region of the plateau is so deeply dissected that were the water not detained by a vegetal covering it would flow off after a heavy rainfall before it had time to enter the limestone, and the streams would have such volume and velocity as to cause swift and destructive floods. The Edwards Plateau is not covered with continuous forests even in the most favorable situations. The timber is much interrupted by open eg ~ es g iS SS Fig. 163. — Escarpment timber of the Edwards Plateau. Conral River near its source. (U. S. Bur. of For.) grassy uplands. Tongues of luxuriant forest follow the stream-ways into the center of the limestone region and in the deeper, well-watered, sheltered canyons the trees attain large dimensions. The forest is in the form of a thick-canopied, shady cover protecting many shade-loving shrubs. This floral community is altogether unlike that found in the ad- jacent country. It is distinctly like the Atlantic type, from which it is separated by miles of treeless country. The chief representatives are the American elm, sycamore, pecan, cottonwood, walnut, black cherry, etc., and in some places cypress with a diameter of 5 feet or more and a corre- 438 FOREST PHYSIOGRAPHY sponding height. The timbered belt is confined rather strictly to the deeply eroded portions of the plateau, gradually giving way to prairie on the level uplands toward the west and to grassy plains toward the east. There is a gradual dwarfing and thinning out of the heavy timber as one passes from the generally heavy growth of the watered canyons to the stunted forests of the hills and bluffs and the still scan- tier tree growth of the loose stony slopes. Finally there remain only scattered chaparral and the vegetation of the low Sotol Country, Fig. 157, whose principal representatives are sotol, cactus, yucca, and agave. The hill and bluff forest occurs on the “ breaks ’’-of the Colorade along the escarpment front from Austin westward and on the Guadalupe, the Pedronalles, and the Freio. It also extends northward upon the breaks of the Grand Prairie and the jagged hills of the granite country. The timber of this type varies in density with local conditions. On lower flats where deep black soil occurs there is a heavy mixed growth of cedar, live oak, elm, hackberry, mountain oak, shin oak, etc., and the type is extended to the side gorges and draws leading from the main stream-ways. ‘Ten miles northwest of Austin on the Colorado are some timber-capped buttes, and similar occurrences are found in a few other localities. On the unstable talus slopes, where the natural shortage of water is emphasized by the excessive porosity of the loose débris, no timber covering is found except a scattered growth of mountain cedar.! Juniper and laurel occur along the vertical slopes of the scarps and sometimes encircle the hills with bands of evergreen. On the floors of the drier valleys are tongue-like extensions of the chaparral flora of the Rio Grande Valley. They are characterized by thorny deciduous trees, mostly acacias, with an undergrowth of Mexican nopal. Thick-skinned yucca, ixtle, and cacti are also found as true desert flora on the bare limestone of the numerous buttes about the borders of the plateau. On the broken limestone areas are the “oak shinneries,” marked by a predominating growth of oak or dwarf shin oak. These are simply dense thickets, scarcely more than tall shrubbery, as on the divides be- tween the Colorado and the San Gabriel drainage country in Burnett County, where much of the growth is so dense as to make it impossible to ride through on horseback. Though it has no value as timber it is a soil retainer of great value. Mixed with this growth isa certain amount of live oak, mountain oak, plum, sumach, holly, and a number of climbers.” 1 W. L. Bray, The Timber of the Edwards Plateau of Texas: its Relation to Climate, Water Supply, and Soil, Bull. U. S. Bur. For. No. 49, 1904, pp. 15-17. 2 Idem, pp. 17, 18. PLATEAUS AND RANGES OF THE PLAINS COUNTRY 439 Among the most valuable assets of the Edwards Plateau region are the cedar brakes conspicuous on the white arid hills of crumbly limestone where the cedar is the dominant and practically the only species. It also grows in mixture with other species, attaining its largest growth in the mixed forests of the lower uplands, where the water supply and soil con- ditions are better. The most extensive bodies of cedar are those of the Colorado River brakes from Austin to the San Saba country.!. The cedar brakes are dry and as likely to burn asa prairie of tall grass, Evidences of ancient or recent fires are found almost everywhere. Many of the.trees of the Edwards Plateau region are peculiar prod- ucts of their environment. “The eastern red cedar here becomes the mountain cedar; black walnut is represented by the Mexican walnut, whose nuts are tiny balls scarcely half an inch in diameter. Texas oak becomes mountain oak. The common live oak becomes a new form in its mountain habitat. The common persimmon is represented by the Mexican persimmon, whose fruit is a dark blue-black; Canadian redbud is here also a characteristic ‘Judas-tree,’ but of a different species. The same is true in the case of the buckeye, mulberry, hackberry, and still others.’’2 The type of vegetation in the Edwards Plateau is undergoing a transition from grass to woody growth. The mesquite is capturing the open pastures, and the scrub oak occupying uplands that were for- merly open prairies. The transition is due to a number of causes. The ridges have been over-pastured and the balance of power thus thrown to the shrubs which are the vanguard of a timber covering. Twenty-five years ago the prairie held sway over large areas where now one finds scrub oak on every side. A great deal of the “shin- nery country”? undoubtedly represents a recent gain in timber on the prairie divides; from the edge of the brush each year new sprouts or seedlings are pushed out a few feet farther, and the new growths soon offer shelter for others. These scattered vanguards were formerly killed by the prairie fires and the timber growth held in check; but with the spread of settlements prairie fires have been reduced and a means of holding its position withdrawn from the grass covering. One of the most striking aspects of this encroachment of the timber upon the prairie land has been the spread of mesquite over the cattle country. Pastures have often been covered with a thicket as close as that of scrub oak, in which, however, the mesquite forms an open orchard-like growth with which are finally associated various species of chaparral and very commonly the prickly pear and the Opuntia. The final result of the spreading of mesquite is a heavy covering of vege- 1 W. L. Bray, The Timber of the Edwards Plateau of Texas: its Relation to Climate, Water Supply, and Soil, Bull. U. S. Bur. For. No. 49, 1994, p. 19. 2 Idem, p. 15. 440 FOREST PHYSIOGRAPHY tation which serves as a protector of water supply and soil, although the original grassy growth would be of far greater value to man.! Some of the interrelations of forests, water supply, temperature, and soils are peculiarly interesting. The forest covering furnishes shelter to the ground beneath from the sun’s rays and prevents intense heating of the rocks and soils. A reduction of temperature is also afforded by the constant transpiration of water vapor from the leafage of the forest. A far more important effect, however, is the influence of the forest upon the soils. The foliage breaks the force of the rain and compels it to run harmlessly down the trunk or to drip slowly through the leaves. The direct impact of the falling raindrops upon the soil is thus pre- vented, erosion greatly diminished, and the absorption of the rain water promoted. Tending to the same result is the organic material of the forest floor, which holds back the fallen water until it has had time to soak into the soil. Thus gullying is prevented and frequent and destructive floods reduced in number and violence. These results are beneficial not only to the inhabitant of the region but also to those whose welfare is in any degree bound up with the régime of the streams. The rice planter on the coast is just as eager as the ranchman of the plains to have a constant and large flow of water, and the ranchman of the hills wishes his soils preserved and the soil moisture retained. For these purposes a forest cover, especially in a semi-arid region like the Edwards Plateau, is a vital necessity.” Brack HILis The Black Hills of southwestern South Dakota and adjacent por- tions of Nebraska and Wyoming rise several thousand feet above the 77 Iss Fig. 164. — Ideal east-west section across the Black Hills. Vertical scale six times the horizontal. 1, slates and schists; 2, granite; 3, sandstone (Potsdam); 4, limestone (Carb.); 5, sandstone (Trias.); 6, shales (Jura); 7, shales (Cret.); 8, shales (Tertiary). (After Newton.) surface of the surrounding plains. They have, by reason of their ele- vation, an abundant rainfall and as a result are well wooded, have many streams, and are in effect an oasis in a semi-arid region. They are 1 W. L. Bray, The Timber of the Edwards Plateau of Texas: its Relation to Climate, Water Supply, and Soil, Bull. U. S. Bur. For. No. 49, 1904, pp. 23, 24. 2 Idem, pp. 26, 27, 20. PLATEAUS AND RANGES OF THE PLAINS COUNTRY 441 carved from a dome-shaped uplift of the earth’s crust; the uplift has been sufficient to cause vigorous erosion, so that the upper layers once forming the top of the dome, and corresponding in age and charac- ter with the strata on the surface of the plains, have been removed by erosion, and the sedimentary rocks now exposed on some of the moun- tain summits of the Black Hills are older than those forming the surface of the Great Plains. The length of the Black Hills dome is about too miles, the width 50 miles. The chief features are (1) a central area of Tig. 165. Western slope of Black Hills southeast of Newcastle, Wyoming looking southeast. Steep- dipping beds are limestone, which spread out in a plateau at foot of slope. (Darton, U.S. Geol. Surv.) high ridges culminating in Harney Peak (7216 feet). This central area is composed of intrusive crystalline schists and granite of several varieties. (2) About the central crystalline area occur various concentric rings of sedimentary rock separated by well-developed valleys; the innermost mass of sedimentary rock occurs in the form of a limestone plateau with an infacing escarpment, Fig. 164. The limestone plateau slopes outward and extends around the Black 442 FOREST PHYSIOGRAPHY Hills. Near its base there is a low ridge of limestone with a steep infacing escarpment from 4o to 50 feet high. The escarpments and slopes are sharply notched here and there by canyons which form char- acteristic “gates.” Between this limestone plateau and the hogback ridge which constitutes the outer rim of the Hills is a depression, the Red Valley, which extends continuously about the uplift. The Red Valley is in many places two miles wide, though it is much narrower where the strata dip more steeply. It is one of the most conspicuous features of the region and is called the Red Valley on account of the red color of its soil. The outer hogback rim presents a steep face toward the Red Valley, above which it rises several hundred feet; on the outer side it slopes more or less steeply down to the plains that encircle the hills. It is crossed by numerous gaps and canyons which divide it into subordinate ridges of various lengths. The Black Hills dome was first developed either in early Tertiary or in late Cretaceous time, but the first uplift was to a moderate height. After this first uplift the larger topographic outlines of the region were established, the dome truncated, and its largest encircling valley ex- cavated in part to its present depth. Later deposits (Oligocene) were laid down by streams and in local lakes, and these deposits extend far up the flanks of the Black Hills, as at Lead and Bear Lodge moun- tains, respectively. After the deposition of these beds the Black Hills dome was raised several hundred feet higher and more extensively eroded, an erosion that has continued through Quaternary time and is In progress to-day.! SOILS AND FORESTS The soils of the Black Hills region are closely related to the under- lying rocks and are of residual origin except on the valley floors and in a few cases where eolian action has taken place. On the limestone plateaus calcareous material forms the greater part of the rock. The soluble portions have been removed and the insoluble portions of clay and sand have collected as a soil mantle which varies in thickness with the character of the limestone, being thin where the limestone is pure, and thick where it contains many impurities. The amount of soil in a given region depends also on the rate of erosion. On many slopes erosion removes the soil as soon as it is formed, leaving bare rock 1 Darton notes the occurrence of outliers of Tertiary (Oligocene and Miocene) deposits high up on the slopes of the Black Hills (N. H. Darton, Geology and Underground Water of South Dakota, Water-Supply Paper U. S. Geol. Surv. No. 227, 1909, pp. 26-27). PLATEAUS AND RANGES OF THE PLAINS COUNTRY 443 surfaces; on others the surface is flat, erosion is not active, and the soil has accumulated to considerable depth. In the central core of crystalline schists and granites the rocks have been decomposed most by the dehydration of portions of their feldspar, with the result that the derived soil is usually a mixture of clay, quartz grains, mica, and other materials. The shales yield a sandy soil where they are sandy and a clayey soil where they themselves have been formed of relatively pure clay. In many cases the geologic forma- tions alternate in short distances; there are corresponding abrupt tran- sitions in the character of the soil in narrow parallel zones. Lack of sympathy between soils and underlying rock may be seen in the river bottoms, in sand dunes, in areas of high river gravels, and on slopes where soil derived by slope wash are mingled with or covered by soils derived in place. The Black Hills forest in general terminates abruptly at the broad valley known as the ‘Race Track,” which lies between the main mass of the uplift and the lines of ridges which encircle the hills. The higher portions of the encircling ridges are also clothed with forest. Sometimes the growth in the latter case is good, sometimes it is in the form of a narrow summit fringe of trees. The main portion of the Black Hills forest. includes about 2000 square miles of densely timbered territory. In many places the continuity of the forest is broken by parks and mountain prairies, and large tracts have been destroyed by forest fires which have swept the Black Hills periodically for years if not for centuries. The yellow pine is the only species of commercial importance; the others are either too small or have too specialized a use to have any great value. A few small bodies of spruce occur, and aspen has come in on some of the burned tracts. The best-quality pine is found in the side ravines and canyon bottoms, where soil and water supply are most favorable and where protection is afforded by the topography. On the steep slopes the soil is stony and thin, the drainage excessive, and the trees shorter and smaller. The forests on the north slopes are in gen- eral better than those on the south slopes because of (1) better pro- tection from fire and (2) greater water supply on account of lessened insolation.! In general the limestone soils are more fertile than those derived from other kinds of rock and bear the most vigorous growths of trees. 1 H.S. Graves, Black Hills Forest Reserve, roth Ann. Rept. U.S. Geol. Surv., pt. 5, 1897-08, PP. 72-75. 444 FOREST PHYSIOGRAPHY OUTLYING DoMES On the northern border of the Black Hills the Great Plains are diver- sified by a number of picturesque elevations that are unique topo- graphic features besides affording illustrations of a very peculiar type Fig. 166. — Devil’s Tower from the north. The steep laccolithic mass is phonolite; the gentler slopes about it are developed on shales and sandstones. (Darton, U.S. Geol. Surv.) of structure. Each hill owes its existence to the injection from below of a column of molten rock into stratified beds. They are not to be considered as volcanic necks, which they in some respects resemble, for the injected rock did not reach the surface so as to form either coulees or cinder cones. Named in general order from east to west PLATEAUS AND RANGES OF THE PLAINS COUNTRY 445 the principal hills are Bear Butte, Custer Peak, Terry Peak, Black Butte, Crow Peak, in South Dakota, and the Inyan Kara, the Sun Dance Hills, Warren Peaks, Mato Tepee or the Devil’s Tower, and the Little Missouri Buttes in Wyoming. As a group these hills display variety in the degree of erosion and hence in the degree of preservation of the original structures. In some cases the dome of stratified beds covering the injected rock is still un- broken and the plutonic rock concealed; in others erosion has exposed the plutonic core, while in some erosion has progressed to the point where the central core stands forth as a tower of columnar rock (phonolite) several hundred feet in height. The first type is illustrated by Little Sun Dance dome, which has very regular outlines, is about a mile in diameter, and deeply scored by erosion but not sufficiently eroded to expose the core of igneous rock that presumably lies underneath.! The eroded type is illustrated by Mato Tepee, in which the arch of sedimentary rock has been entirely removed, exposing a column of injected rock. The tower stands on the west bank of the Belle Fourche River, quite by itself, and rises to a height of 626 feet above the plat- form on which it stands. The shaft of this magnificent natural column is composed of cluster prisms which extend from base to summit with- out cross divisions, each prism being a uniform unbroken column more than 500 feet high. Since the plateau on which Mato Tepee stands is itself about 500 feet above the river, it is clear that the minimum amount of erosion that must have taken place to expose the column is somewhat over tooo feet. It was probably much over this amount, for none of the material of the Butte was extruded at the surface. There is good reason for believing that the amount of erosion in the region is but little under three-fourths of a mile, a conclusion which means that this great tower must have been buried under at least a half mile of sedimentary rock. LittLtE BELT, H1GHWoOOD, AND LITTLE Rocky MOUNTAINS In central Montana, just east of the front ranges of the Rocky Moun- tains, are a number of isolated mountain groups with structural and topo- graphic qualities totally different from those of the northern Rockies. On the east they break the continuity of the Great Plains of Montana and are prominent topographic features long before the traveler sights the Rocky Mountains. Lying east of or on the eastern border of the main Cordillera and in a region of deficient rainfall, their own sum- 17. C. Russell, Igneous Intrusions in the Neighborhood of the Black Hills of Dakota, Jour. Geol., vol. 4, 1896, pp. 23-43. 446 FOREST PHYSIOGRAPHY mits rise to such heights, 5000 to gooo feet, as to induce a greater rainfall than occurs on the surrounding plains, and they are therefore clothed with forests. Among these mountain groups are the Little Belt, High- wood, Little Rocky Mountains, and others. LirrLtE Bett MOUNTAINS The Little Belt Mountains are an elevated and eroded plateau, as shown on the accompanying topographic map, Fig 167. They are about 60 miles wide from east to west and 40 miles from north to south on the west side, and taper to a sharp point at Judith Gap on the east, so that the group is roughly triangular in shape. Individual peaks along the northeastern border of the group rise above the general level and form an uneven crest line visible from the open plains. As compared with the well-defined ranges of the Rockies the Little Belt group is rela- tively low and wide and composed of many spurs radiating from a central point.1 The border of the mountains on the west is deflned by the deep canyon of Smith River, beyond which the country has gentler relief. On the summit of the Little Belt arch the rocks are gently inclined or horizontal, but on the flanks or shoulders of the arch the rock dips steeply away from the uplift. The intrusions which arched the beds are laccolithic in character. Since the development of its structural features the range has suffered extensive denudation. Erosion has laid bare the larger laccoliths and worn down the general level differ- entially; the harder igneous rocks have been left in relief to form the higher summits, while the softer sedimentary rocks have been eroded, although the members of each group have been ercded at very different rates owing to differences in degree of resistance. On account of. deficient height the Little Belt Mountains did not support local ice sheets during the glacial period, nor were they covered by the continental ice sheet during the time of its maximum southward extension. Throughout the greater part of the Little Belt region a plateau-like topography prevails; broad flat summits are characteristic. The aver- age elevation is 7600 feet, though the summit level from which the spurs radiate is 8000 feet high. The highest summit of the group is not at the center but along the northeastern border, where Big Baldy reaches an altitude of gooo feet. The Little Belt Mountains are 1 W. H. Weed, The Geology of the Little Belt Mountains, Mont., 20th Ann. Rept. U. S. Geol. Surv., pt. 3, 1898-1809, p. 273. 2 Idem, p. 277. PLATEAUS AND RANGES OF THE PLAINS COUNTRY 447 ZI \ \ San V), } SH Rk Us s) Me ee mG + NS 3 Fig. 167.— Topographic map of the Little Belt Mountains. (U.S. Geol. Surv.) 448 FOREST PHYSIOGRAPHY bounded by relatively soft rocks to whose deep erosion the prominence of the mountains is chiefly due. The relation of the detailed topographic features to the rock character is so intimate that it is impossible to distinguish the one feature without distinguishing the other. Summit plateaus are bordered by steep es- carpments with towering limestone cliffs along the stream gorges. Only the highest peaks along the northeastern border and the steep limestone gorges lend picturesqueness to the scenery. In the center of the mountain area where the beds are horizontal or only gently in- clined, secondary structural plateaus are commonly found. Upon some of these the rock is so resistant as to determine broad ridges which in some cases are emphasized by differences in soil and vegetation, as is the case in Belt Park and other timberless parks near Neihart, which have been developed on quartzite and which are in contrast with the wooded and soil-covered slopes above and below. There are no broad valleys within the mountains; for the most part the streams flow in deep trenches or narrow canyons in the harder limestone and in less narrow valleys in the shale belts.! CLIMATE, SOIL, AND VEGETATION Both rainfall and snowfall are relatively abundant in the Little Belt Mountains. Intermittent streams which flow only in wet weather or at times of melting snow are common. ‘These are especially character- istic in limestone areas where the waters are absorbed by the porous and fissured rock and in those regions where the catchment areas are small. As a consequence of the greater precipitation due to the greater height of the Little Belt Mountains over the surrounding plains they are in general forest-clad, their dark slopes being in strong contrast to the arid treeless plains about them. Lodgepole pine is the prevailing species; In some places it forms forests with individual trees 10 to 40 inches in diameter, but usually it is much smaller. On the plateau summits the white pine is found, and spruce and fir grow along the wet stream bottoms and on moist and cold northern exposures. There is a very intimate relation between the character of the tree growth and the nature of the slope exposure. On southward-facing slopes, which are relatively dry because of the greater insolation, the growth is sparse and open, and interrupted by grassy parks; on north- ward-facing slopes thick and dark forests occur. The growth varies somewhat also with the character of the rock and the physical nature 1 W. H. Weed, Geology of the Little Belt Mountains, Mont., zoth Ann. Rept: U. S. Geol. Surv., pt. 3, 1898-09, p. 275. PLATEAUS AND RANGES OF THE PLAINS COUNTRY 449 of the derived soil. The shales produce but little soil and support scanty vegetation; they usually underlie the park regions. The sand- stones and the igneous rocks are covered with land waste and are gen- erally densely wooded. HIGHWOOD MOUNTAINS The Highwood Mountains are a prominent member of the group of elevations that break the continuity of the northern Great Plains. They lie in the great bend of the Missouri in central Montana about 20 miles north of the Belt Mountains, and about 50 miles southwest of the Bear- paw Mountains, Fig. 148. About them stretch the smooth monotonous plains above which they rise about 4ooo feet to summit elevations of 7600 feet. The Highwoods are a group of old and now much eroded volcanoes which broke through the once flat-lying Cretaceous strata of the plains. The mountains are composed chiefly of volcanic flows and breccias and a number of stocks or central cores, the remnants of former more lofty cones. The outer foothills are low and rounded; toward the center the country becomes more rugged. The descent to the plains is abrupt on the south; the chief feature of the northern slopes is an old glacial spillway known as Shonkin Sag whose origin is explained on p. 416. The drainage is in the main radial and consequent upon the original slopes of the volcanoes. The streams are rather constant in flow in the mountains but become sluggish and alkaline on the plains, some of them drying up in summer to such an extent that water remains only in pools in the deepest portions of the stream channels. The outer foothills and valley openings have been occupied by ranchmen who also utilize the water for limited irrigation. Extensive pastures are found on the higher slopes, though these give way to true forests of small pines on the northern exposures. In many places there are dense thickets of lodge- pole pine. The name ‘“ Highwoods” undoubtedly had its origin in the forest growth of the northern slopes.! LITTLE ROCKY MOUNTAINS The Little Rocky Mountains lie about 200 miles east of the Rocky Mountain Cordillera and between the Missouri and the Milk about 60 miles south of the 49th parallel. They rise from 2000 to 3000 feet above the treeless plains of central Montana, forming a conspicuous topographic feature in a plains region that is generally without 1L. V. Pirsson, Petrography and Geology of the Igneous Rocks of the Highwood Moun- tains, Montana, Bull. U. S. Geol. Surv. No. 237, 1905, pp. 1-22. FOREST PHYSIOGRAPHY 450 (CAINS "Joan 'S “AQ ‘Lez “[[NG) 799} Coe “PeAIOJUL IMOJUOD “LURJUOPL ‘suTeJUNOJ pooMyYsiAZ oy} jo deur s1ydessodoy, — gor “sry TEN J Ny i me ) a ageey Nv i WNSKG xh 3.090 117) 4 ASN Ca Wag 2IAASOd ening Ooo CS Cis D = Oy, G n b9.q)/S | », A ya Y iG ¢ y} Y QO ny ») t C 4 wg yy Y y JS y 0 PLATEAUS AND RANGES OF THE PLAINS COUNTRY 451 prominent landmarks. Their topographic prominence led the Indians to call them ‘‘Eah hea Wwetan,” or the “Island Mountains.” The Little Rocky Mountains have an undulating crest without sharp peaks. The highest summit is more than 6500 feet above the sea, though only about half that height above the surrounding plains. The scenery is attractive but not grand, for the mountain summits are generally rounded, without that boldness that is the striking feature of most alpine scenery; and to this softness of outline is added the soften- ing effect of a thick growth of small pines, which covers almost all of the summits. Perhaps the most picturesque aspects of the mountains are developed about their borders, where heavily bedded limestones (Car- boniferous) are cut by deep narrow canyons variegated by a vegetation that stands in pleasing contrast to the dry plains. The limestones form a white wall encircling the mountains, and stream and cliff erosion has cut the thoroughly jointed beds into huge white scarps visible from points 50 miles away. Within the mountains the streams flow in deep V-shaped gorges. These mountains are formed upon a single dome-shaped uplift having a central core of crystalline schists overlain and marginally wrapped about by limestones (Paleozoic) and softer beds (Mesozoic).' Hard and soft layers encircle and overlap the mountains and are crossed -one after the other by streams consequent upon the original slopes of the uplift. Subsequent streams have developed along the strike of the softer beds and usually join the consequents where the valleys of the latter are broadest. The central core of crystalline schist is exposed in the headwater gorges of all the larger streams and in the deep-cut side slopes of the main crest. The gorges have heavily timbered slopes; the prevailing type of timber is the lodgepole pine, from 3 to 20 feet high. It is characteristic of the granite to have a covering of young pines; pines are found also upon the débris slopes, while the limestones and schist areas are covered by the big-leaf pine, which forms groves alternating with open parks.” OzARK PROVINCE The approximate limits of the Ozark region are the Missouri and Osage rivers on the north, the Arkansas on the south, the Neosho on the west, and the Black River on the east; the Shawnee Hills of southern 1 Weed and Pirsson, Geology of the Little Rocky Mountains, Jour. Geol., vol. 4, 1806, pp. 399-428. 2 Tdem, p. 410. 452 FOREST PHYSIOGRAPHY Illinois, a continuation of the Ozark plateau, extend eastward to Shaw- neetown on the Ohio River. The Ozark region may be described as a broad, relatively flat-topped dome somewhat extensively dissected and consisting of three sub- divisions, the Salem Platform, the Springfield Structural Plain, and the Boston Mountains. The Springfield Plain inclines at a low angle toward the west in Missouri and toward the southwest in Arkansas, a slope which corresponds with the dip of the underlying formation. It Saleme sv eae SALEM Mise ys eae oe, gi ) -- TEAU. Tenth 55k. VAN LEY R Gu laf /ovacwira MTNS. Atoka t : me Fig. 169. — Subdivisions of the Ozark region and relations to surrounding provinces. (Taff, U. S. Geol. Surv.) is deeply dissected by the large streams which flow through it in narrow valleys; the interfluves are large tracts of broad flat structural surfaces from which younger formations have been eroded. The Boston Mountains are capped by thin layers of sandstone and shale, the more resistant sandstone governing the physical features of the mountains. The rocks dip in the main to the south at an angle slightly greater than the general southward slope of the surface. From hilltops on the Springfield Plain the Boston Mountains appear as a bold, even escarpment with a level crest, but on closer examination the escarpment is seen to have many finger-like extensions to the north PLATEAUS AND RANGES OF THE PLAINS COUNTRY 453 in the form of ridges and foothills. Toward the west the Boston Mountains decline in elevation and in ruggedness as the sandstone beds become thinner and more shaly in that direction. The Boston Mountains on the southern border of the Ozark province owe their dominating height in part to the excessive erosion of the region north of them and in part to differential uplift.' The result of erosion ay SCALE 6° 10° 20-30 40 ‘50 'OOMILES ® Fig. 170. — Topography of the Ozark region. (Purdue, U.S. Geol. Surv.) was to reduce a large part of the Ozark region in Missouri and Arkansas to a comparatively low altitude, leaving the Boston Mountains as re- siduals and their front as a rather bold escarpment. The former greater extent of the Boston Mountains is indicated by remnantal outliers standing several hundred feet above the general level of the area about them. In contrast to the northern part of the Ozark region, which is a low flat dome with only local faulting and minor undulations, the 1 A. H. Purdue, Winslow Folio U. S. Geol. Surv. No. 154, p. 5, col. 4. 454 FOREST PHYSIOGRAPHY Boston Mountains have a monoclinal structure and a correspondingly steeper border topography. The topographic forms of the Ozark region are those characteristic of early maturity in a region of nearly flat-bedded rocks of varying hardness in which the dip of the rock and the slope of the surface are in many places coincident. The entire northern slope of the uplift is a succession of broad flat plateaus separated by more or less ragged es- carpments which mark the margins of the harder formations. For- ward from the escarpments there are scattered outliers, the fragments of more extensive layers separated from the escarpments by circum- erosion. The greatest dissection of the region is on the south and east, the least on the north. On the south and especially at the eastern end of the Boston Mountains the topography is rougher and the plateau character is ali but lost. On the north the streams run commonly through broad and rather shallow valleys between which lie extensive areas of relatively undissected plateau. The drainage is in the main radial and consequent, the stream directions corresponding with the OZARK PLATEAU ST. FRANCIS MTNS. a ee SEA LEVEL Fig. 171. — Topographic and structural section across the Ozark region along line A—A in Fig. 170. Shows the crystalline rocks of the St. Francis Mountains, the limestones of the Ozark Plateau, and the shales, sandstones, and limestones of the Boston Mountains. restored slopes of the structural dome, but a certain amount of subse- quent stream development has also taken place where tributaries have excavated valleys along the outcrop of the softer formations. Seen from commanding points the surface of the Ozark region presents the appearance of an almost unbroken plateau due in part to the flat- ness of the structure but also and in larger part to peneplanation that once brought diverse structures to a common level which the vigorous erosion of the region since uplift has not yet wholly destroyed. The two main subdivisions of the Ozark Plateau, the Salem upland and the Springfield Plain, are separated by the Burlington escarpment, which runs in a general north-south direction.!| The escarpment is formed upon the border of a limestone layer (Mississippian) which con- stitutes the surface rock of the western part of the plateau. East of the escarpment, outliers of the limestone occur in the form of small residual areas between which are dissected plains bearing chert accu- mulations derived from the weathering and erosion of the limestone 1 C, F. Marbut, Physical Features of Missouri, Missouri Geol. Surv., vol. 10, 1896, pp. 11- IIo. PLATEAUS AND RANGES OF THE PLAINS COUNTRY 455 that once extended over the entire region. The streams of the Salem upland flow in narrow valleys some of which are 250 feet or more deep. The degree of dissection is in some places so great that the former ex- tensive plain is not easily recognizable, though large interstream tracts - still preserve their plain-like character. The Springfield upland on the other hand is largely a structural plain developed on the surface of the Mississippian limestone. Near the east- ern border of the upland the streams flow in shallow trough-like valleys which deepen westward as the streams cross low anticlines and domes. Those that drain the upland eastward have deep valleys where they cross the edge of the Mississippian limestone, which they have dissected to a ragged fringe. , SOILS AND TREE GROWTH The soil of the Salem upland is cherty to a high degree, the chert having been derived from the overlying limestone. On weathering, the chert breaks into angular blocks and, because of its greater durability, forms a surface layer of débris. The finer soil particles produced by the weathering of the limestone are carried downward to the base of the weathered zone and are therefore at too great a depth to be accessible to agriculture, though they are of first importance to the forests, which seem to thrive in spite of the surface accumulations of loose stones. The most luxuriant forest is formed upon soils derived from the light-blue limestone and blue shales of the Morrow formation (Carboniferous). Walnut, locust, and other types found naturally only on fertile soils occur on this formation in spite of the fact that it out- crops usually on steep slopes. In general it may be said that in spite of the rather poor soils of the entire region, due to the cherty residual products resulting from the erosion of the limestone and also to the considerable dissection which the region is undergoing to-day, the sur- face is rather well occupied by forests, since it is too stony and steep to serve for other purposes and since the finer soil is washed down to too great a depth below the surface to be available for the ordinary plants of agriculture. ARKANSAS VALLEY South of the Boston Mountains is the Arkansas Valley district, which is structurally much more complex than the Ozark region, although standing at a lower elevation. The underlying strata have been thrown into overlapping folds which have been beveled off by erosion so com- pletely as to form a local peneplain approximately 800 feet above the 456 FOREST PHYSIOGRAPHY sea. It is surmounted, however, by residual peaks and ridges of small area and from 1700 to 2500 feet above sea level. Since the formation of the Arkansas Valley peneplain, uplift has occurred and the soft shaly beds have been worn away, leaving the inclined sandstones as low, narrow, and sharp-crested ridges whose summits are generally horizontal, a condition indicating the former elevation and topographic character of the area. OvacHItaA MouNTAINS The Ouachita Mountains lie south of the Arkansas Valley and extend from Little Rock, Arkansas, westward into eastern Oklahoma. The range is 200 miles long and nowhere rises more than a few thousand feet above the surrounding valleys and plains. The topographic forms of the range are developed upon an Appalachian type of structure, p. 585. Near the center of the range, long wide folds are developed upon massive sandstone; on the borders of the uplift are shorter and more complex folds developed upon shales and limestones as well as sand- stones.!. The regular development of hills or ridges and valleys is a striking feature of the group as contrasted with the irregularities of the igneous knobs and ridges of the Wichita Mountains of western Okla- homa. The regularity of the topographic details is absent in many places along the border, however, where faults with vertical displace- ments of several thousand feet have combined with the overturning of © the folds to make the relief more irregular both in general plan and in detail. ARBUCKLE MOUNTAINS The Arbuckle Mountains form a triangular area approximately 30 miles on each side with a westward extension. The western part of the Arbuckle Mountains has an elevation of about 1300 feet or 1400 feet above the plains on either side. The mountains were uplifted (late Carboniferous) and base-leveled in common with the great Appalachian province east of the Mississippi. Upon the base-leveled surface, then an almost flat plain, Cretaceous deposits were laid down. Later uplift caused the removal of the Cretaceous strata over large areas, but the superior hardness of the underlying rock has resulted in its preserva- tion in the form of a low plateau in the central part of the mountain group, while the bordering strata have been eroded to a lower level. The minor topographic details of the Arbuckle Mountains are due chiefly to the varying resistances of the rock formations. The lime- 1 J. A. Taff, Structural Features of the Ouachita Mountain Range in Indian Territory, Science, n. s., vol. 11, 1900, pp. 187-188. PLATEAUS AND RANGES OF THE PLAINS COUNTRY 457 stones in general are hard and form level-topped and narrow ridges with crests approximating the general plane of the regional uplift. Interven- ing soft cherts and shales are found in the wooded valleys that separate the ridges. In the more elevated part of the uplift erosion by swift streams has etched the border of the plateau into a frill of deep gulches. Fig. 172. — Mixed hardwoods, etc., in typical relation to topography, Arbuckle Mountains, Oklahoma. On the dry limestone formations forests are absent, except along the lower and moister valley slopes and the valley floors. (Reeds, Oklahoma Geol. Surv.) A second and lower plain (probably Tertiary) occurs in the Wichita- Arbuckle region and descends approximately with the grade of the rivers. This plain merges into the Tertiary deposits of the coast on the one hand, and on the other hand stretches westward into Oklahoma and northward across Indian Territory. It is preserved through eastern and southeastern Oklahoma in innumerable ridges and hills, occurring at elevations approximating 1800 feet. A third erosion surface has been identified in the plains surrounding the Arbuckle Mountains; the larger streams have cut wide and flat valleys 200 feet below the gen- eral level of the Tertiary peneplain. ‘This erosion surface is, however, much more fragmentary than either of the two surfaces just described. It is found bordering the Ouachita Mountains in the form of wide flat valleys developed upon the softer rocks.! Scrub Oak, Red Cedar, and Red Bud are the common trees on the mountain slopes, while red and white elm, Bois d’Arc or Osage orange, 1 J. A. Taff, Preliminary Report on the Geology of the Arbuckle and Wichita Mountains in Indian Territory and Oklahoma, Prof. Paper U.S. Geol. Surv. No. 31, 1904, pp. 15-17. 458 FOREST PHYSIOGRAPHY hickory, pecan, hackberry, honey locust, river plum, papaw, mulberry, sycamore, willow, and cottonwood grow in the creek valleys and on the flood plains of the rivers like the Washita. There is also a striking increase in the timber growth (1) on the weaker formations (Sylvan and Simpson) which weather to lower levels, and (2) on those formations (Woodford and Reagan) which contain appreciable amounts of phos- phate and iron. The chert, shale, and sandstone strata, together with the granite and porphyry of the East and West Timbered Hills, are all covered with timber. In the narrow creek valleys and canyons, which have been developed on limestones (Arbuckle, Viola, and Hunton), trees appear on the alluvial deposits and near slopes, but rarely on the up- land surface, Fig. 172. The rainfall is notably heavier in the mountains than on the adjacent plains. Where the slopes are cleared for culti- vation the soil washes so badly that the clearings are soon abandoned. While the clearings are not abundant, yet in the aggregate they include an important and always a conspicuous part of the total area! WicH1ITA MOoUNTAINS The Wichita Mountains are the westernmost of the three mountain groups of the southern Great Plains. They are composed of igneous rock, Fig. 173. — Border topography, Wichita Mountains, Oklahoma. (Gould, Oklahoma Geol. Surv.) chiefly granite, and have been eroded into peaks varying in height from a few hundred feet to 1500 feet above the surrounding plains. 1 The data for this paragraph have been supplied by Prof. C. A. Reed of Bryn Mawr College and Prof. C. A. Gould of the University of Oklahoma. PLATEAUS AND RANGES OF THE PLAINS COUNTRY 459 The main range is about 30 miles long and 12 miles wide. West of it are a number oi scattered smaller ranges and peaks, such as Mount Tepee, Quartz Mountain, and Headwater Mountain. North and east is a 30-mile parallel range of hills composed chiefly of hard massive lime- stone, and the same limestone on the south and east occurs in the form of small rounded knobs. The limestone hills represent remnants of a series of rocks which once extended as a dome over the igneous rocks but which have since been deeply eroded. Between the granite and limestone ridges on the borders of the uplift are softer “Red Beds,” shale and sandstone with local deposits of conglomerate, and these have been eroded to form valley lowlands whose extent corresponds to the outcrop of the less resistant strata.! Mounts Scott and Baker are the highest mountains in the group and, like the numerous other peaks in the region, they have steep, rugged, bowlder-strewn slopes. Although the relative altitudes of peaks, ridges, and valleys are small, the mountains have a distinctly rugged appear- ance, with sharp outlines and narrow passes. There are more than 250 detached areas of igneous rocks in the Wichita Mountains group, and these are expressed topographically in the form of a main mountain mass 150 square miles in extent and a large number of neighboring isolated sharp knobs which rise like islands above the smooth plains about them. The whole group forms an archipelago of granite moun- tains and peaks rising rather abruptly from the sea-like plains devel- oped upon softer rock.” 1 C. N. Gould, Geology and Water Resources of Oklahoma, Water-Supply Paper U. S. Geol. Surv. No. 148, 1905, p. 15; J. A. Taff, Prof. Paper U. S. Geol. Surv. No. 31, 1904. 2 J. A. Taff, Preliminary Report on the Geology of the Arbuckle and Wichita Mountains in Indian Territory and Oklahoma, Prof. Paper U. S. Geol. Surv. No. 31, 1904, pp. 54, 77. CHAPTER XXIV PRAIRIE PLAINS EXTENT AND CHARACTERISTICS BETWEEN the great Laurentian area of Canada on the north and the Ozark-Appalachian provinces on the south and extending east and west from the Great Plains to the Appalachian Plateaus is a broad expanse of moderately rolling country known as the Prairies or Prairie Plains. The province includes the greater part of the Middle West, but it is not confined to that section of the country. Its western por- tions are thinly timbered, the forest growth shading off to scattered groves of timber in lower and wetter localities or to the valley floors and the stream margins. Its eastern, better-watered portions are covered with a denser arboreal growth, though clearings are so exten- sive as to leave but insignificant patches of the primeval forest, and everywhere the timber is confined chiefly to wood-lots of limited and generally decreasing extent. The timber of the southern two-thirds of the Prairie Plains is prevailingly hardwood; on the north is a belt of coniferous forest which fills the gap between the northern edge of the hardwoods and the spruce forest belt of Canada. Although the Prairie Plains support a dense agricultural population and supply a disproportionately large part of the corn, wheat, oats, etc., of the country, they do not have an exceptionally favorable rainfall. It is certain that the agricultural products are far below the possibilities of the soil were a heavier and better distributed rainfall to occur. There is good reason for believing that were the surface less flat, the absorp- tion of rain water less pronounced, the original forest would have been much more restricted. The Prairie Plains belong to the vast central plains region between the Atlantic and the Pacific Cordillera and present from commanding points a number of characteristic views. In many places the province has the appearance of a limitless expanse of grove-dotted, gently undu- lating country, here and there trenched by rivers and surmounted by low hills; or it stretches away as far as the eye can see without either of these relief features —a grass-covered, farm-dotted, smoothly contoured prairie. North of the Missouri and Ohio rivers it is glaciated. Morainic belts cross it in looped pattern, and between them are notably flat till plains. 460 PRAIRIE PLAINS 401 In that portion covered by glacial ice in the last (Wisconsin) glacial in- vasion, lakes, ponds, and undrained hollows in great numbers are scattered freely about, stream courses are disorganized, falls and rapids abound and alternate with swampy depressions often of considerable extent. The southern portion of the glaciated tract and a part of the unglaciated area beyond are covered to a variable degree by loess deposits, — fine, > Fig. 174. — Typical view of the Prairie Plains in the Great Lake region. Woodland tract in left foreground. light, wind- and stream-deposited detritus of great importance in relation to both run-off and soil fertility. East of the Mississippi the Prairie Plains grade southward into the Appalachian Plateaus; west of that river and south of the limit of glaciation they have a distinctive quality unlike any other portion of the province. In the Osage Prairie of eastern Kansas, and specifically about Independence, they have the appearance of an ancient surface of erosion. Here broad valleys separated by low divides are characteristic features; none of the irregularities of glacial topography or drainage is found; in fact there is so little relief that water storage by dam construction is impracticable on account of the breadth of the valleys and their gentle gradients. 462 FOREST PHYSIOGRAPHY THE PENEPLAIN OF THE PRAIRIES The most general topographic feature of the Prairie Plains is the Tertiary peneplain now standing at various elevations above its former level. It may be traced from central Texas northward through Okla- homa and Kansas into Wisconsin, Indiana, and southern Michigan. While it is now a dissected and therefore a discontinuous surface its for- mer character and wide extent may reasonably be inferred because large tracts are still in an excellent state of preservation. It is fair to assume that the Tertiary peneplain of the region was far more extensive than its visible remnants, now for the most part in process of dissection. The Tertiary peneplain appears to have been much more extensive in the upper Mississippi basin than was the corresponding plain in the Appa- lachian tract; the latter was limited largely to the belts of softer rock or to the larger streams. The uplifted peneplain was dissected to various stages of maturity before the Pleistocene period, so that by the beginning of that period pronounced valleys had been formed: in some places the valleys weré narrow and the intervening divides wide, while in other places the val- leys were widely opened and the divides reduced to narrow ridges. The slopes of many valleys were steep and covered with a thin veneer of decayed rock, and almost the whole of the now glaciated country had a somewhat more rugged appearance than at present. The surface of the peneplain was not everywhere affected in the same way after its development. In some cases it was uplifted and broad valleys opened at lower levels, in other cases there appears to have been no uplift, in still others uplift appears to have been progressive and no opportunity given for the formation of a peneplain. Even with- out peneplanation the surface would nowhere have great relief, for this is distinctly a region of low-lying and nearly flat strata. The depth of erosion conditioned by the rainfall, the elevation, and the character of the rock would be a measure of the relief, unlike many mountain regions where the relief in the earliest stages of an erosion cycle is commonly related directly to original structural irregularities. A description of a few typical occurrences will serve to show the general development of the peneplain of the prairies and its present condition. An ancient topographic level has been clearly determined in southwestern Wisconsin (Driftless Area). Fragments of the plain slope gently southward across the outcropping edges of shales, dolomites, and sandstones. In spite of the strong variations in resistance to erosion which these beveled strata display the plain truncates them with PRAIRIE PLAINS 463 striking uniformity. It is therefore not a structural plain but a plain of erosion. It is typically developed about Lancaster, Wisconsin, where it stands about 1100 feet above the sea. Since its formation it has been broadly uplifted and extensively dissected. The main valley bottoms lie several hundred feet below the general level of the uplifted peneplain and well-developed flood plains are common. Dissection has progressed to the point where the original plain surface now remains only along the stream divides, and the cubic contents of the valleys approximately equal the cubic contents of the ridges between the valleys.1 South of the Great Lake region the uplifted and dissected peneplain may be clearly identified. In eastern Missouri and in Illinois it has been described as a plain of wide extent now much dissected, its remnants being represented by the accordant hill and ridgetops. In a general view this regularity of level of the summit areas makes the sky line nearly flat, a dominating topographic feature. Such striking accordance of summit levels is the more significant when it is realized that the plane thus denoted truncates inclined strata offering very unequal resistances to erosion.” In Indiana the surface appears to display two erosion levels forming an upper and a lower plain. The level summits of the higher hills and ridges mark the position of the now greatly dissected older topographic level about 600 feet above the sea. One hundred feet lower are rolling uplands which exhibit smooth, gently rounded hills and ridges or flat divides separated by broad and relatively shallow valleys.’ In the prairie plains of eastern Kansas the uplift appears to have been insufficient to occasion dissection of the peneplain formed during the Tertiary. The beveling of the strata is here still in progress. The divides are smooth and broad, truncation is now in progress; if the land retains its present level there will be even more complete leveling of the surface, and the region will finally exhibit an ultimate base-leveled plain. All relation between structure and topography has, however, not yet vanished. In the Osage Prairie, for example, are broad structural terraces surmounted by a few outlying and isolated hills, though the relief is far less gentle and faint as compared with the relief of the region during the earlier periods of the present erosion cycle. The terraces face eastward and have frontal scarps from too to 200 feet 1 Grant and Burchard, Lancaster-Mineral Point Folio U. S. Geol. Surv. No. 145, 1907, p. 2, cols. 1-2. 2 N. M. Fenneman, Physiography of the St. Louis Area, Bull. Ill. Geol. Surv. No. 12, 1909, PP. 17, 52. 3 Fuller and Clapp, Ditney Folio U. S. Geol. Surv. No. 84, 1902, p. 1, col. 3. 464 FOREST PHYSIOGRAPHY high. Each terrace has a long and relatively gentle westward slope whose total descent is less than the descent of the frontal scarp. Each terrace therefore stands above its neighbor on the east and below the next terrace on the west, so that the series rises in westward “succession and the rate is about 10 feet per mile. The north-south drainage lines show a marked tendency to migrate down the dip of the individuai terrace, i.e., westward; hence the eastern valley slopes are commonly gentle and low, the western, steep and high. Stream courses at right angles to the terrace fronts show alternations of broad open stretches on the terrace tops and narrow steep-sided stretches on the terrace fronts. Both the topographic and the drainage features are related to the structure in still other respects. The westward-dipping shales, sand- no on = =| o °o 2 2 oS 2 cS = = c 3 r= = 2 ° = = o = =) a 5 © ates: 5 ° a = = eK) = = ® 5 op) 50 Ghee = ae Oo f S et og Yn ie r ie < (oye =} ° = Gig Cat S00) Ss ° ° Sis a a oO. shaft es) a el = =o E ae E E 2° Bios oo S) an 3) oO oO nA O 6) LEED 5. SAILS Orel x<> DONT AE Fig. 175. — North-south section near Tishomingo, Oklahoma, showing the peneplaned surface of the Prairie Plains on the south. (Taff, U. S. Geol. Surv.) stones, and limestones are of variable thickness. Where the resist- ant sandstones are thick the topography is irregular and has marked relief; where the soft shales are thick and the resistant strata thin or absent the surface has very gentle gradients, broad valleys, and but little relief. The average elevation of the district is about goo feet, and the difference in elevation between adjacent hills and valleys is from 50, t0.200 feet 4 Toward the south, as in Oklahoma and northern Texas, the Prairie Plains once more take on the character of an uplifted and dissected peneplain, Fig. 175. The uplift was not great, however, probably not more than 1oo feet, and large remnants of the former featureless plain occur on the interfluves. The valley topography is a feature of the present cycle of erosion; the inter-valley topography exhibits many characters inherited from the preceding cycle of erosion in which a peneplain was formed.? 1 F. C. Schrader, Independence Folio U. S. Geol. Surv. No. 159, 1908, pp. I-5. 2 J. A. Taff, Coalgate Folio U. S. Geol. Surv. No. 74, 1901, p. 4, col. 4., and C. N. Gould, Geology and Water Resources of Oklahoma, Water-Supply Paper U. S. Geol. Surv. No. 148, 1905, PD. 12. PRAIRIE PLAINS 465 CENTERS OF GLACIATION ! The ice sheets which overrode the northern Prairie Plains originated in two centers of glaciation known as the Keewatin, west of Hudson Bay, and the Labrador, in the Labrador peninsula. A third, the Cordilleran center in northern British Columbia, is of little interest here, for it lay too remotely to the west to affect to an important degree » (e3 Zz SoS ») (i Mi (eee L 73, ” et ro) ates) {{(((( eee eet) (WArx v { thgnroogy \ ey) WwwSS a, he SS f= 40° be i | 10)° Fig. a6 —- The centers of ice accumulation, the position of the Driftless Area, and the southern limit of glaciation. (Alden, U. S. Geol. Surv.) the topography of the Plains, Fig. 176. The Keewatin glacier formed in a great gathering ground close to the western coast of Hudson Bay, from which the ice radiated in all directions — eastward into Hudson Bay, northward to the Arctic Ocean, westward to the Mackenzie Val- ley, and southward toward Manitoba and the Great Plains. At or near the center of glaciation the striz are very indefinite and appear to have changed in direction as the center slightly shifted its position, but no evidence of more than one glaciation has been found, nor has it been 1 J. B. Tyrrell, The Genesis of Lake Agassiz, Jour. Geol., vol. 4, 1896, pp. 811-815. 466 FOREST PHYSIOGRAPHY determined that the ice left the country uncovered at any time during the glacial epoch. The Keewatin glacier advanced southward and southwestward until it came in contact with the high escarpment of Cretaceous~shales in eastern Manitoba, followed the course of Lake Winnipeg and Red River and in this direction advanced far into Minnesota, Dakota, and Iowa. The eastern margin of this lobe did not extend very far east of the present eastern shore of Lake Winnipeg, and it is probable that throughout its advance there was a free drainage eastward, probably into Hudson Bay. The evidences of the time of advance of the three continental ice caps which formed over northern North America go to show that these three seem to have reached their widest extent and to have retired in suc- cession from west to east, and.that a fourth ice cap, probably similar in character to those that have disappeared from the American continent, covers Greenland at the present time. GLACIAL AND INTERGLACIAL PERIODS The different glacial and interglacial stages of the glacial period are numbered in the order of their occurrence below.! Their deposits are represented in part on the accompanying maps showing relations of the drift sheets in the Middle West. XIII. The Champlain substage (marine). XII. The glacio-lacustrine substage. XI. The Later Wisconsin, the sixth advance. X. The fifth interval of deglaciation, as yet unnamed. IX. The Earlier Wisconsin, the fifth invasion. VII. The Peorian, the fourth interglacial interval. Vil. The Iowan, the fourth invasion. VI. The Sangamon, the third interglacial interval. V. The Illinoian, the third invasion. IV. The Yarmouth or Buchanan, the second interglacial interval. III. The Kansan, orsecond invasion now recognized. II. The Aftonian, the first known interglacial interval. I. The sub-Aftonian, Jerseyan, or Nebraskan, the earliest known invasion. I. The sub-Aftonian or Nebraskan ? glacial stage is represented in Iowa as a very old drift sheet whose surface bears sand and gravel, peat, 1 Chamberlin and Salisbury, Geology, vol. 3, 1906, p. 382 ff. 2 B. Shimek, Bull. Geol. Soc. Am., vol. 20, 1909, p. 408. PRAIRIE PLAINS 467 old soil, and other products of prolonged weathering. The Nebraskan drift consists largely of compact blue-black bowlder clay. In many places it is thickly set with woody material gathered from forests that were overwhelmed by the ice. The wood is largely spruce, cedar, or conifer- ous species that indicate a cool-temperature flora in advance of gla- ciation. The drift filled the deep preglacial valleys to depths of nearly 350 feet, but it formed only a thin cover on the uplands, where it was nearly removed by erosion before the next glacial invasion.! The interglacial stages are of little interest in the present connection. Rather typical conditions are represented by the first epoch in the series. II. The Aftonian interglacial stage is represented by sand and gravel deposits and by beds of peat and muck, with stumps and branches of trees, and shows prolonged erosion and weather- ing. Several bones of the camel have been found in the deposits of this epoch; also the antler and fragments of bones of a large stag, three species of elephant, the American mastodon, and a large variety of molluscan remains of both aquatic and terrestrial species. The fauna and flora indicate a climate not greatly different from the present, though the fossils merely represent the conditions at the time of their burial, and still leave largely to conjecture the full range of temperature of the period.’ III. The Kansan glacial deposits, p. 473, occupy a large area in Kansas, Missouri, Iowa, and Nebraska and probably extend under the later glacial formations far to northward. The Kansan till is pro- nouncedly clayey and has very little water-laid material either within the body of the till or on its margin, where the usual fluvio-glacial deposits of sand and gravel are very meager. The surface of the Kansan till although originally quite flat has been considerably eroded and its topographic features notably altered from their original expression. It is estimated that the Kansan drift occupies only from 10% to 30% of the original plain. The streams flow in valleys with broad and low (3° to 5°) slopes and bottoms. The average drift removed by erosion in the exposed portion of the Kansas drift is estimated to be not less than 50 feet. The material is deeply weathered and it has been deprived of its finer and more readily dissolved calcareous material and even limestone pebbles down to 5 or 8 feet.3 V. The Illinoian till is chiefly clay and is similar to the Kansan till in showing a marked absence of assorted drift in most places. Drain- age modification brought about through the Illinoian glacial invasion in the Mississippi Valley is of uncommon interest. The western edge of the Illinoian ice lobe crossed the Mississippi Valley between Rock Island and Fort Madison, and pushed the Mississippi River out of its course a score of miles (p. 475). 1 Frank Leverett, Comparison of North American and European Glacial Deposits, Zeitsch. f. Gletscherkunde, Berlin, vol. 4, 1910, p. 250. 2 Idem, p. 253. 3 Idem, p. 258. 468 FOREST PHYSIOGRAPHY The bowlder clay of the Illinoian drift appears originally to have been very calcareous but to have been leached since deposition to a depth of from 5 to 8 feet and nearly all the limestone pebbles completely removed. Much of the deposit has become partially cemented because of the large calcareous element in the drift. Where erosion has been most rapid it has destroyed scarcely more than half the original plain. 5 VII. The Iowan till is thin and is noteworthy for the exceptional number of large granitoid bowlders which lie chiefly on the surface. This drift was formed by a lobe of the Keewatin ice sheet. The moraines formed about the borders of the lobe were very feeble and the outwash scant. IX. The two Wisconsin ice sheets are the most important of all, not because of their greater original extent or their thicker drift deposits but because their drift deposits are the last of the series and have there- fore been more extensively preserved. The interval since the occurrence of these last glacial stages has been so brief, furthermore, that but little modification of topography, drainage, or soils, has been possible. The outermost limits of the Wisconsin ice sheets are marked in most places by pronounced terminal moraines, and in addition the surfaces of the till sheets are diversified by terminal moraines due to the periodic recession of the ice. Their surfaces are also diversified by various gla- cial and fluvio-glacial forms not developed, as a rule, on the deposits of the earlier glacial stages. The chief varieties of these deposits are the out- wash plains formed by streams discharging from the ice front, also kames, eskers, and drumlins. In a number of noteworthy cases these forms have a marked tendency toward aggregation. The Sun Prairie topographic sheet of Wisconsin represents the remarkable drumlin accumulations of that and adjacent portions of the state; and a similar swarm of drumlins may be seen in central New York between the Finger Lake district and Lake Ontario. The drumlins are oriented in the direction of ice movement and are composed of unstratified and very compact till? The two glacial advances of Wisconsin time were separated by a rather short interval of deglaciation, and the readvance in the second of the two stages was along slightly different lines, so that the relative size and relations of the ice lobes appear to have undergone consider- 1 Frank Leverett, Comparison of Nerth American and European Glacial Deposits, Zeitsch. f. Gletscherkunde, Berlin, vol. 4, 1910, p. 270. 2 For discussion of drumlins or, more properly, rocdrumlins, formed out of shale and earlier till deposits in Menominee County, Michigan, and that drumlins are forms of aggradation in some cases and of degradation in others, see I. C. Russell, The Surface Geology of Portions of Menominee, Dickinson, and Iron Counties, Michigan, Rept. Geol: Surv. of Mich. for r906, and H. L. Fairchild, Drumlins of Central Western New York, Bul]. New York State Mus. No. 111, 1907, Pp. 393 et al. PRAIRIE PLAINS 409 able change, and the moraines of the later stage cross those of the earlier at distinct angles, as at some points on Long Island (p. 470). The drift sheet of the later of the two stages is characterized by enormous terminal moraines, great bowlder belts, and an unusual development of kames, eskers, drumlins, outwash plains, valley trains, ete., all formed in such a manner as to show the disposal of the ice in the form of great lobes which in turn reflect the influence of the basins in which they were formed. The late Wisconsin drift is in many notable instances thin and overlies the interglacial fluvial deposits as a thin veneer of till. TopoGRAPHIC, DRAINAGE, AND SoIL EFFECTS TOPCGRAPHY In the successive periods of glaciation the ice advanced over the up- lifted and dissected peneplain of the northern Prairie Plains described in the preceding pages. In places it increased the relief by irregular deposition on a flat plain, in other places it decreased it by deposition in the valleys of a dissected plain. Had the country been mountainous an overriding ice sheet would not have produced till plains of such topo- graphic uniformity and continuity. The present relief is therefore a function not only of glacial but also of preglacial topographic form. The topographic and drainage effects of the successive ice invasions are of the first magnitude in the xegion in which they occur, while the distribution of the soils is almost everywhere related to the material which the ice laid down or which the draining streams swept forward in the region immediately beyond that covered by the ice. The southern limit of ice advance is indicated upon the map, Plate , but it must not be thought that a single vast sheet of ice at any one time occupied the entire region north of the line indicating the limit of glaciation. There were in all at least six ice sheets which swept from the north at as many different times, from two of the three recognized centers of ice accumulation. The deposits produced by the successive ice invasions occur as distinct drift sheets between which, in intervals of deglaciation, soils and beds of peat and marl were formed (p. 467). In addition to these lines of demarcation between successive till sheets there are differ- -ences among the sheets themselves; they vary in extent and physical constitution in a marked way. The lines of flow in successive invasions of ice were in many cases far from coincident, so that ice fields in some places fell short of earlier ones and in other places exceeded them. Another important feature is the existence of pronounced morainic ridges along the fronts of some of the till sheets, although this is not a 470 FOREST PHYSIOGRAPHY constant feature, since a wide stretch of country in Illinois, Kansas, and Iowa is marked by the absence of a terminal moraine. Where the moraines are present it is noteworthy that the earlier ones have been very extensively eroded, so that many irregularities found upon the later moraines are wanting in the earlier. A typical occurrence is found in southern Illinois east of St. Louis, where practically all the kettle holes and minor depressions of the Ilinoian moraine have been either filled up, or drained by the cutting down of their outlets, or both, while the detailed irregularities of the slopes of the hills and ridges have been so completely smoothed and rounded as to appear to a certain extent artificial. This group of qualities is in marked contrast to the numerous sags, depres- sions, kettle holes with lakes or swamps on their floors, and the extreme rugosity of moraines formed during the last (Wisconsin) ice advance. TERMINAL MORAINES AND TILL SHEETS The terminal moraines marking the margin of the last Wisconsin invasion are pronounced features of the landscape and often rise from 100 to 200 feet above the sur- rounding surface. They may be traced almost continuously west- ward from Long Island to and beyond the Great Lakes. Not all the ice invasions of previous gla- cial epochs resulted in the devel- opment of pronounced terminal moraines. Some of the till sheets Pa : He have distinct marginal ridges, way tA while others are marked by mere i fet | swells of thicker drift at the mar- gins, and still others have very thin edges. The marginal deposits of the Wisconsin drift are in the form of bold terminal and interlobate moraines of such topographic importance that they have served to modify or to form entirely the divides between the smalier rivers. They fill a number of preéxisting river channels and deflect many of the small streams and even many larger ones from their former courses. The concentric trend of the morainic ridges is to be ascribed to the Fig. 177. — The four drift sheets of Wisconsin. (Weid- man, Wisc. Geol. Surv.) PRAIRIE PLAINS 471 lobate character of the ice margin, a feature imposed upon the conti- nental ice sheet by the irregularities of the topography which it covered. The depressions appear to have been lines of maximum flow, and in the Fig. 178. — Distribution of glacial moraines and direction of ice movement in southern Michigan and northern Ohio and Indiana. (Russell and Leverett, U.S. Geol. Surv.) Great Lake region this resulted in the strongly looped quality of the morainic systems as shown in Fig. 178. The extent to which drift deposits dominate the glacial topography is further indicated by the fact that the present divide between the Great Lakes or St. Lawrence drainage and the Mississippi drainage is 472 FOREST PHYSIOGRAPHY determined very largely by moraines and thick drift deposits. The drift is heaped up in greatest amounts at the ends and along the sides of the Great Lake basins where it was accumulated on the margins of tongues or lobes of ice that once occupied the basins. Such modifi- cation of not only the drainage but also the topography and the soils is clearly indicated by the nature of the drift covering in the south- Driftl i eae riftless Fs) Moraines it inotan Ty lowan drift 753] Leess and sand ridges on lowan LT drift border Shelbyville till sheet J (Early Wisconsin) Champaign till sheet (Early Wisconsin) Bloomington till sheet (Early Wisconsin ) Post- Bloomingten drift Fig. 179. — Generalized glacial map of northern Illinois. (Barrows, Bull. 15, Ill. Geol. Surv.) ern peninsula of Michigan, northwestern Ohio, northern Indiana, and northeastern Illinois. The till deposits here attain their greatest thickness. The southern peninsula of Michigan is quite remarkable in this respect since the geographic position and relations of the Great Lake basins determined a maximum convergence of the ice lobes of this district. The northern half of the peninsula has a drift cover from 700 to 800 feet thick; the surface rises in places to 1000 or PRAIRIE PLAINS 473 1100 feet above the surface of the lakes; there seems to be no rock more than 250 feet above the lakes.' It has been estimated that the entire southern pen- = insula of Michigan has an aver- | ue age of about 300 feet of drift. ARs ee ros The drift cover in the adjoining g ibe pee states on the south is about roo 3 4 SA Sy sg feet thick. These heavy glacial e Hy fe $e Rs é deposits do not consist of asingle |ewr3din7 Sle a5 he sheet, but embrace three, and in a ns Heels wg places four, distinct drift sheets | ssi | : {® which are the products of repeated [§ | | ' & bl om glaciations at widely separatedin- |? / }- 45 vanca | esr) & \ I tervals. The upper surfaces of k \ \ poy ‘ the older sheets are marked by NE eee 2 5 i beds of peat and river and lake | ON Lie deposits containing remains of she life forms similar to those NOW Fig. 180. — Positions of the Wisconsin ice lobes about existing in the region, an indica- the Driftless Area. (Weidman, Wisc. Geol, Surv.) ¢ 5 See tion of a temperate interglacial climate not markedly unlike the climate of to-day. Farther west, as on the plains of eastern North Dakota, are many _AVISCONSIN/ Lie, Fig. 181. — Relations of the drift sheets of Iowa and northern Illinois. 1, Kansan; 2, Illinoian; 3, Iowan; 4, Early Wisconsin; 5, Late Wisconsin; 6, Driftless area; 7, course of the Mississippi River during the Illinoian glacial epoch. (After Leverett, U.S. Geol. Surv., and Calvin, Iowa Geol. Surv.) 1 W. F. Cooper, Water-Supply Paper U.S. Geol. Surv. No. 182, 1908, Plate IT. 474 FOREST PHYSIOGRAPHY surface features which also show the characteristic effects of glaciation. The country is here for the most part level, but it also presents long rolling slopes rising from 300 to 800 feet above broad valleys. The most important topographic elements are massive ridges or mesas due to preglacial erosion. The mesas are in many instances bordered or crowned by long morainic ridges representing halts in the glacial advance or retreat. The morainic material is derived chiefly fromthe underlying shales (Cretaceous), and is a compact clay which also contains erratic fragments and bowlders of crystalline rock derived from northerly localities. The upper 5 to ro feet of till has been oxidized to a light yellow color. The intermorainic tracts consist of rolling plains of till from a fraction of an inch to a hundred feet thick or of more level plains due to alluviation on the floors of glacial lakes. A typical in- stance of lake-bed topography is sk \ S found in the upper James River Fig. 182. — Southern limit of the Pleistocene ice sheet and the valley of the Red River.! and distribution of moraines of the Dakota glacial 3 i J lobe, North and South Dakota. (Willard, U. Ss. Postglacial stream dissection has Geol. Surv.) further diversified the topography, the Missouri River having cut a trench several hundred feet deep whose sides are for the most part steep. SO aPed Poo OTANI d-—eee ——. ee “-"XGettysburg 5 2 Preryiue 7 ( oS \ } 5, os, is < 2) S DRAINAGE MODIFICATIONS One of the interesting effects of glaciation was the diversion of streams from northward to southward courses; an instance of this is the Mis- sissippl in the southwestern corner of the Driftless Area below Dubuque. The valley maintains an even width of 1} miles for about 12 miles, then widens gradually, and in a stretch of about 60 miles its width in- creases, except for a slight local contraction near the mouth of the Wisconsin River, and finally becomes 3? miles. This decided widening of a low-gradient valley upstream suggests that in preglacial time the valley was occupied by a northward-flowing river. i J. E. Todd, Aberdeen-Redfield Folio U. S. Geol. Surv. No. 165, 1900, pp. 7 et al. PRAIRIE PLAINS 475 Other cases of stream diversions are discussed in connection with the Appalachian Plateaus, page 685, the Great Plains, page 4o5, and the Connecticut Valley, page 638. More general and quite as con- spicuous are the influences of glaciation upon the drainage sys- tems laid out upon the surfaces of the various till sheets or dis- posed along their margins. The Ohio and the Missouri have courses in marked sympathy with the glaciated tract, as if they had been pushed bodily out of pre- glacial courses, and such indeed is the case at many points on both Ds aa Gee 5 ES Erie #4 WY, streams. The Ohio in particular 2 s 4 exhibits a valley whose most strik- |spencer Leclaire @ / ing characteristics are open-broad Sy stretches alternating with rock- walled gorges which closely con- fine the stream; and abandoned channel stretches are exhibited at many points. Many less con- spicuous illustrations of stream diversions along the border of the glaciated country have been discovered! Within the borders of the till sheets influences no less striking are found. The moraine crests are the chief minor divides of the glaciated country and in many cases they are also the major divides. Number- less tributaries starting on the slopes of moraines are gathered into trunk streams which flow for long distances along the margin of a moraine or between parallel moraines before finding an outlet across one or the other of the moraines into a transverse stream (pp. 471 and 479). Stream courses within the morainic belts are characteristically irregular and marked by many lake-like expansions alternating with steep stretches and often by falls and rapids of great economic value. These conditions are in sharp contrast to the regular drainage lines and the general absence of lakes and waterfalls in the streams draining the smooth outwash plains that in many cases front the moraines. DY ., a > or dove 1S 22g by Fig. 183. — Old and new channels of the Mississippi at the upper rapids, Fulton, Ill. 1 Frank Leverett, The Illinois Glacial Lobe, Mon. U.S. Geol. Surv., vol. 38, 1899, pp. 102- 103 et al. 476 FOREST PHYSIOGRAPHY The preglacial drainage of the Great Lake region was probably very unlike the drainage of to-day, a for borings along the line of the buried channel that runs across Michigan from Saginaw Bay to Lake Michigan indicate a fall toward the south- west and also a widening of the channel in that ee aung direction. Similar deep borings in western In- diana between the head of Lake Michigan and the Wabash show a lower rock floor than any yet discovered across Illinois, and suggest a preglacial drainage to the southward toward the Wabash and Ohio.! It seems probable that there was a divide on the line of the present St. Lawrence below Lake Ontario where the river now flows among the Thousand Islands from which the drainage was southwestward. The outlines of all the lake basins have been notably affected by the deposition of drift in part on their margins, in part on their floors. Near the mouth of the Cuyahoga River at Cleveland the drift extends more than 470 feet below lake level and to within too feet of sea level, while the greatest depth of water of Lake Erie is but 210 feet and the average depth only about_ 70 feet. These figures indicate a very notable modification of the western end of Lake Erie by the deposition of till, and such modification when brought about by a large ice mass must in- evitably be associated with very notable changes in the preglacial drainage. Fig. 184 is a profile across Lake Michigan, showing the extent to which drift forms a barrier to the lake waters at the present time. Drift 179 feet Sandstone 76 feet Pleistocene 0-Carboniferous (Leverett, U. S. Geol. Surv.) Surface of lake 579.6 feet Bed of lake Sea Level LAKE PLAINS The so-called Lake Plains of the Great Lake region are special features of the Prairie Plains province. They have an aggregate extent of several thousand square miles and form long narrow strips of smooth country bordering the Fig. 184. — Profile across Lake Michigan from Milwaukee to Grand Haven, showing the relation of glacial deposits to the extent of the lake. aaynvom |W (a) 1 Frank Leverett, Outline of History of the Great Lakes, 12th Rept. Mich. Acad. Sci., 1910, p. 22. PRAIRIE PLAINS 477 Great Lakes in sharp contrast to the rough morainic country about them. Many portions of them have low gradients, smooth sur- faces, regular consequent drainage, fine soils, and extensive marshy tracts that remind one strongly of portions of the Atlantic and Gulf Coastal Plain. They have an aggregate extent of several thousand square miles, but this is far less than the area usually assigned to them. There is no doubt that their extent as represented on Powell’s physio- graphic map of the United States! is greatly exaggerated. For the most part they are closely confined to the borders of the Great Lakes. In places they extend inland for 50 or 60 miles, but such breadths are excep- tional. Commonly their width is under 20 or 30 miles. The Lake Plains lie between morainic ridges and the lake shores. The morainic ridges are concentric with respect to the existing lakes since the lake basins were lines of movement of the glacial lobes along whose margins the moraines were developed. The lakeward slope of the inner- most moraine formed the border of the jake and against this slope beach ridges and wave-cut platforms were developed in many cases. Where the innermost moraine was feebly developed or lower than its outer neighbors it did not determine the lake border. Furthermore, the lakes once lying over the lake plains were constantly changing in level and hence in outline. Shore lines are found between the outermost abandoned shore line and the present lake shores. These were developed not upon moraines but upon smooth lake floors uncovered by the retreat of the lake waters as they fell to lower levels. The shallowness of the glacial-marginal lakes and the gradual move- ments of the shore zone of maximum wave action over the entire lake floor exercised a smoothing effect upon the lake bottom elevations. These lake bottoms, now exposed, in many cases exhibit a topography varied only by recent stream channels opened in them and by shallow depressions filled with small swamps, ponds, and lakes. PROGLACIAL LAKES” It was known even to the first settlers that lakes formerly occupied not only the present lake basins but also the bordering lands thus over- lying the present lake plains and forming the prominent beaches developed 1 J. W. Powell, Physiographic Regions of the United States, in Physiography of the United States, Nat. Geog. Mon.. 1806, pp. 98-00. 2 Whittlesey and Warren, The Great Ice Dams of Lakes Maumee, Am. Geol., vol. 24, 1899, pp. 6-38. See Frank Leverett, The Glacial Formations and Drainage Features of the Erie and Ohio Basins, Mon. U. S. Geol. Surv., vol. 41, 1902, for an excellent brief historical review of the problems of the abandoned shore lines of the Great Lake region as well as for a compact summary of the physical changes and a good working bibliography. See also J. W. 478 FOREST PHYSIOGRAPHY in numbers about the existing lakes at both high and low levels. Many of the beaches once served as trails for the Indians. Their level, dry, and moderately straight courses are also ideal for modern purposes. They are generally known as “‘ ridge roads” and are fairly common in the southern part of the Great Lake region. It was early demonstrated! that the lakes owed their origin to ice dams now extinct, and that the water impounded in front of the ice dams rose to the level of the lowest point on the rims of the enclosing basins and then overflowed into valleys draining to the sea. The ice dams were nothing more or less than the frontal lobes of the retreating continental ice cap of late glacial time. : To understand these relations one should conceive of a great continen- tal ice sheet whose front was being melted back from a southern limit, Plate IV, so that it finally lay on the northern side of the great divide _ which separates the St. Lawrence drainage system from the Mississippi and Atlantic drainage. From the pattern of the terminal moraines shown in Figs. 178 and 182, the border of the continental ice sheet is known to have been lobate. Each lake basin as well as each subsidiary bay, such as Green Bay or Saginaw Bay, was occupied by an individual lobe of ice connected at the north and east with the main ice sheet. In front of each ice lobe and in the southwestern and western portions of the St. Lawrence basin there eventually formed a glacial-marginal or proglacial lake which was at first crescent shaped, but with continued retreat of the ice its shape would be modified by the general outlines of that part of the basin that was not occupied by the ice lobe. It is also clear that the lake water could not discharge over the thick ice sheet or under it, but must have discharged along its front or across the divides on the distal margins of the basin. Lake Chicago is the name applied to the crescent-shaped lake that formed in front of the Lake Michigan lobe and discharged through the Desplaines-[linois rivers. In a simi- lar way Lake Maumee was formed in front of the Erie lobe and dis- Goldthwait, The Abandoned Shore Lines of Eastern Wisconsin, Bull. Wisconsin Geol. and Nat. Hist. Surv. No. 17, 1907, for historical data of interest and for unusually accurate and detailed studies of abandoned shore lines in Wisconsin. More recently observation has been extended into Canada by Taylor and Goldthwait and the conclusions based on studies in the United States correlated with the observations of Coleman and others on the Canadian Geo- logical Survey. The results of the refined studies of recent years point conclusively to the truth of Spencer’s early contentions that the deforming movement of the region was differ- ential and greater toward the north, a conclusion long neglected outright or regarded with undeserved skepticism chiefly because of the author’s retention of the overthrown hypothesis of marine submergence. 1G. K. Gilbert, On Certain Glacial and Postglacial Phenomena of the Maumee Valley, Am. Jour. Sci., vol. 1, 3d ser., 1871, pp. 339-345. PRAIRIE PLAINS 479 E. Fig. 185. — Drainage history of the southern Great Lake district. MAG position of glacial lakes before lakes were formed; B, beginning of Lake Chicago; C, D, E, F, later stages in retreat of ice and expansion of proglacial lakes; G, retreat of ice from Mohawk valley and discharge of proglacial lakes down the Hudson; H, present drainage. charged through what is now the valley of the Maumee into the Wabash past Fort Wayne. Lake Duluth was formed in front of the Superior lobe and discharged down the Chippewa to the foot of Lake Pepin in the Mississippi Valley. On the borders of the glacial-marginal lakes, deltas, shore lines, and 480 FOREST PHYSIOGRAPHY offshore deposits were formed, and on the bottoms, muds, sand, and silts were deposited. One side of each lake had a temporary shore Mtinoig Rive” pe i) pe? SS H Fig. 185. Continued. line — the unstable ice wall which marked the front of each glacial lobe. It must also be noted that this ice wall was not fixed. Some- times it advanced slightly, overriding the beaches that had been formed by the lake water in front of it; sometimes it remained stationary long enough to accumulate a water-laid moraine on the lake floor and a land- PRAIRIE PLAINS 481 laid moraine on the adjacent land surface; but the net result of all the changes of ice front was a gradual retreat northeastward. Asso- ciated with this retreat was the gradual extension of the glacial-marginal lake waters and to a certain degree their coalescence along the ice front, as in the case of the water bodies at the heads of Saginaw Bay and Lake Erie, Fig. 185-E. Numerous changes in outlet also took place, among which only the principal changes are noted below. The principle which controlled these changes is, that with each marked re- treat of the ice lower surfaces were exposed which would have different relations to each other with respect to elevation. Thus Lake Maumee for a time discharged into Lake Chicago by way of the Saginaw basin and the Grand River Valley. Lakes Michigan and Huron for atime discharged through Georgian Bay and the Trent Valley into Lake Iroquois (Ontario), and a still later and lower outlet was afforded through the Ottawa River Valley on the uncovering of this drainage way through the further retreat of the ice. This stage is known as the Nipissing stage. At this time Lake Erie drained across the Niagara escarpment and Lakes Huron, Michigan, and Superior drained eastward through the Ottawa Valley to the St. Lawrence. The small amount of water discharging over Niagara Falls during this period resulted in the production of a very narrow gorge, but the gradual tilting of the land toward the southwest caused the Ottawa outlet to be abandoned and the low col south of Port Huron to be occupied by a new drainage channel marking the outlet of these three great lakes into Lake St. Clair and Lake Erie. Associated with this change in outlet and the greater discharge of water over Niagara Falls was the widening and still more marked deepening of the gorge between the whirlpool and the Suspension Bridge, changes which are conceived to have occurred in the last 5000 years, judging by the present rate of retreat of the falls. The St. Clair outlet of Lake Huron is thus seen to be a very recent geologic event, as may be seen clearly also by the youthful character of the valley of that river and the recent formation of a feature which is anomalous in physical geography — the delta at the head of Lake St. Clair, formed by the short St. Clair River. A river draining a lake is singularly free from sediment at the intake, and it might be expected that a very small delta or none at all would be formed at the mouth of so short a stream as the St. Clair. Instead, the recent occupation of the valley of the St. Clair by the St. Clair River has caused very rapid channel cutting that is still in progress, an action that is reflected in the large delta at the head of Lake St. Clair. Tending toward the same result are the shore currents of Lake Huron, which deliver a large amount of beach sand to the intake of the St. Clair. The former small size of the lake in the Erie basin and the fact that it did not then cover the western part of the basin are established by the deeply submerged lower courses of the streams in the western part of the basin, especially that of Sandusky River, which has been traced entirely across the bed of Sandusky Bay. More recent changes of water level at this end of the basin are shown by the submerged level of the valleys discharging into Lake Erie and the submerged stalactites in the caves of Put-in Bay Island at the western end of the lake. During still later stages of their history the lakes discharged through the Mohawk and down the Hudson. The great sand plain at Albany and Troy was the delta of this large- volume, temporary stream. Through the later uncovering of the St. Lawrence Valley by con- tinued ice melting a lower point of discharge was offered to the lake waters and the Mohawk channel abandoned. The present Mohawk River occupies but a small portion of the ancient glacial channelway. The earliest formation of glacial-marginal lakes in the Lake Superior drainage basin appears to have been in the area now drained by the St. Louis River, where the waters were ponded at 1 L. J. Cole, The St. Clair Delta, Mich. Geol. Surv., 1905. 482 FOREST PHYSIOGRAPHY eae S UTA TOT " & Net TA Fig. 186. — Superior ice lake and glacial marginal Lake Duluth. (Leverett.) Fig. 187. — Lake Duluth at its greatest extent and the contemporary ice border. (Leverett.) PRAIRIE PLAINS 483 an altitude of about 700 feet above Lake Superior, Fig. 186. To this small transient body of water the name of Lake Upham has been given. The fringing ice later afforded a lower passage about 530 feet above Lake Superior, and the resulting lake with outlines quite different from those of Lake Upham and with a different line of discharge has been called Lake St. Louis. Still further retreat of the ice border resulted in the formation of a lower lake about 500 feet above Lake Superior, known as Lake Nemadji, which discharged a large volume of water (from Kettle River Valley) into the St. Croix. These lakes lie quite beyond the border of Lake Superior, and well above it. The name Lake Duluth has been given to the extremely large body of water draining southward from the head of Brule River into the St. Croix River and occupying a large portion of the area now occupied by Lake Superior. The high- est shore line of Lake Duluth at the western end of the basin shows a marked rise toward the northeast. At Duluth it is 465 feet above Lake Superior, while at Calumet on the Keeweenaw peninsula it is 702 feet above it. The rise of 237 feet seems to be largely due to differential uplift, though, as in the case of all the other deformed beaches of the region, a small portion is referable to ice attraction, though this will account for but 37 of the 237 feet and was effective only within a few miles of the ice border. The difference of 200 feet seems explainable only on the ground of differential movements since the formation of the earlier shore lines. During and since the retreat of the ice, tilting of the Great Lake region has been in progress. Elevated and now tilted and abandoned shore lines occur in numbers about the lakes of the Great Lake region. In places they are very numerous, as about the head of Lake Superior, where may be counted more than 30 distinct strand lines, representing as many successive lowerings of water level through the vicissitudes associated with the melting of the ice and the uncovering of lower and lower outlets. The beaches all rise in altitude toward the northeast, with a slightly increasing divergence in that direction, showing that the uplift of the land which gave the shore lines their present attitude was differential and greater toward the north; also that the tilting was not that of a simple plain but of a warped surface.1 With a view to determining the degree of stability of the Great Lake region, Gilbert compared the elevations of certain bench marks as determined in 1876 and again in 1896. After a careful elimination of all irregularities of lake levels due to wind, tide, etc., it was found that there was a small but definite increase in height in a northerly direc- tion, and that the tilting is greatest in a direction south 27° west, and is at the rate of 0.42 foot per 100 miles per century. The prolonged tilting of the Great Lake region is also shown by the drowned lake shores on the west side of the lower peninsula of Michi- gan, where many of the rivers discharge into pouch-shaped lakes sepa- rated from the main lake by long slender sand bars, as the Black River, the Muskegon, the Pentwater, and in a similar way at the western end of Lake Erie, the Maumee, the Rouge, the Raisin, and the Huron. In 1 J. W. Spencer, Deformation of the Algonkian Beach and Birth of Lake Huron, Am. Jour. Sci., vol. 141, 1891, pp. 12-21. 484 FOREST PHYSIOGRAPHY E |e [is 6 G20 ape 611 is 607, Fig. 188. — Isobasic map of the Algonkian and Iroquois beaches, showing the broader features of warping. Numbers indicate elevations of raised beaches above the sea. (Goldthwait.) Fig. 189. — Map of extinct Lake Agassiz and other glacial lakes. (Upham U. S. Geol. Surv.) PRAIRIE PLAINS 485 contrast to these features of shore lines of subsidence are the features indicating elevation on the shore lines of the eastern side of Lake Huron at Kincardine and Goderich, Ontario, where the rising land keeps the lake always at work on new levels, and till, bowlders, cobblestones, and pebbles are everywhere in evidence. The near-shore bottom is stony and the stream discharging into Lake Huron at Kincardine has a stony bed, caving banks, and impetuous flow, indicating uplift and constantly renewed energies.!_ If the tilting continues at the present rate, the water of Lake Michigan will ultimately discharge across the low divide that separates the lake from the Desplaines Valley. Its channel will then be reoccupied and Niagara will become dry. It has been calculated that this change would take place naturally in from 5000 to 10,000 years; but man has anticipated the change by cutting a drainage canal for diverting the sewage of Chicago from Lake Michigan to the Illinois River — a pro- ject enforced by the urgent need of a pure supply of lake water for a concentrated population. The old outlet channels of the glacial-mar- ginal lakes are among the most distinctive topographic and drainage features of the entire region. At the present time the Desplaines has a wide steep-walled valley with a flat marshy floor thronged with lakes, bayous, irregular depressions, and stream channels. The present valley was the channel of the outlet of Lake Chicago, and the irregularities of the present valley are nothing more than the irregularities normal to the channel bottom of a large stream. In the Red River Valley similar drainage changes of equal importance took place. Upon retreating gradually from the Lake Winnipeg-Red River Valley the Keewatin glacier was met by the gradually advancing Laurentide glacier, and the union of the two caused a ponding of the waters between their fronts, which produced glacial-marginal Lake Agas- siz. The lake waters rose until they overflowed southward into the valley of the Mississippi and then gradually declined. The Laurentide glacier advanced to the western shore of Lake Winnipeg; this extreme advance was some time after the advance and retreat of the Keewatin glacier, and between the two till sheets stratified deposits occur near the mouth and on the banks of the Saskatchewan River. The large areas of flat fertile soil in the Red River Valley are largely lacustrine deposits— silts and clays formed on the floor of the ancient glacial- marginal lake and now exposed by the extinction of the lake, Fig. 189. 1 Mark Jefferson, The Geography of Lake Huron at Kincardine, Ontario, Jour. Geog., vol. 2, 1903, Pp. 144-155. 486 FOREST PHYSIOGRAPHY SOILS OF THE GLACIATED COUNTRY Each type of glacial topography in a glaciated country has not only its own distinctive slopes but also its distinctive soils. A map of the glacial forms of a region is therefore to a large extent a soil map of the region. Terminal moraines are noted for stony till, outwash plains. for their porous, gravelly, stratified, alluvial nature, eskers for their cobbly material, and kettle and kame topography is marked by sandy hill- slopes and undrained hollows with soils largely of organic origin. These are not hard and fast criteria but they are in general true. Exceptions to the rule are the sandy moraines near the Great Lakes, the stony out- wash where a late glacial readvance of the ice veneered the surface with a thin ground moraine, etc. Between the concentric terminal moraines. formed during the retreatal stages of glaciation one finds in, many cases extensive tracts of sub-level land formed of till, the so-called ground moraine. These tracts are in many cases poorly drained and their soils. therefore have a high content of organic material. They form the flattest and richest parts of Illinois and Indiana, and lie in the so-called “‘corn belt”’ of these and other states. The unweathered condition of the soil minerals throughout the most. recently glaciated area insures prolonged fertility of the soil. The older glacial deposits have been weathered to an important degree. Both the Kansan and I[llinoian drift sheets, Fig. 181, have been greatly leached and large portions of their original calcareous content removed. The Wis- consin drift is fresh and unaltered, one of its distinguishing features. It is also noteworthy that the older drift sheets have a smaller proportion of organic matter in their soils than is the case with the youngest or Wisconsin sheet. Since the deposition of the Wisconsin drift there has. accumulated large quantities of organic material along the shores of the small lakes and ponds, indeed some of the water bodies have disappeared. through the filling of the basins by sediments, plant remains, and the cut- ting down of the outlets. The older till sheets have been so extensively eroded, on the other hand, that the accumulated organic remains in the small basins have been largely removed and the soils to that extent impoverished. Corresponding contrasts: are exhibited in the value of the soil products, the density of population, and the natural vegetation+ Over the old lake bottoms the soil varies considerably from place to. place according to (1) the nature of the underlying glacial deposits, (2) proximity to a well-defined strand line, (3) character of the post- 1 Frank Leverett, The Illinois Glacial Lobe, Mon. U.S. Geol. Surv., vol. 38, 1800, pp. 734- 736 et al; also An Instance of Geographical Control upon Human Affairs, Geog. Jour. (London), vol. 24, 1904. PRAIRIE PLAINS 487 glacial drainage, etc. In some places the soil is a fine silt, loam, or clay, and is free from stones and bowlders and very fertile. In other locali- ties it is marked by the presence of stones and bowlders, as where a beach was formed against the slopes of a rocky moraine, although for the most part the beaches are composed of sand and gravel. The poorly drained portions of the lake plains are in the aggregate of con- siderable extent and are marked by post-lake accumulations of vege- tation in sufficient amounts to form a surface layer of peat or muck of great fertility when properly aerated by drainage. Many of the most productive of the market gardens that supply the cities of Port Huron, De roit, and Chicago are formed upon these extensive, poorly drained muck swamps. Some of the beach levels are so warm on account of their porous soils and excellent natural drainage as to be ideal sites for vineyards and orchards, a feature well marked in western New York south of Lake Ontario. The lake plains on the eastern side of Lake Michigan and Lake Huron are noted for their large production of fruit, but this development is dependent to a greater degree upon climate than upon soil, for the Great Lakes are exceptionally large bodies of water and mitigate both the heat of summer and the cold of winter. The east- ward or leeward side is favored by sufficiently moderate winter tem- peratures to permit the development of peach, plum, cherry, and apple orchards in great numbers. In the northern part of the southern peninsula of Michigan the glacial deposits are exceptionally sandy and infertile, and large tracts, aggregating nearly one-sixth of the state, are known as the “pine bar- rens.” Six million acres of these lands have been thrown on the delin- quent tax list and are a burden to the people. Formerly they were covered with magnificent forests of white pine. These were so thor- oughly cut down, the surface so extensively and repeatedly burned, and young growth and humus so devastated, that great areas have been rendered all but worthless. Sandy tracts are also found along the lake margins and in some cases, as at the southern end of Lake Michigan, appear to be due to the fact that beach and dune material, formed on the borders of the interglacial Great Lakes, was reworked and redeposited during the last (Wisconsin) glaciation. Further modification has resulted from wind action, the light sands having been blown into dunes some of which attain an enormous size. ‘Creeping Joe” on the eastern shore of Lake Michigan, near Muskegon, is one of the largest dunes in the world, with a relative alti- tude of several hundred feet. The dunes are most numerous on the 488 FOREST PHYSIOGRAPHY eastern shores of the lakes, as on the eastern shores of Lake Michigan and Lake Huron, where sandy deposits are exposed to the full force of the wind sweeping in from the lakes. LOESS DEPOSITS The student of soils must take account of an important deposit in the Mississippi Valley related to glaciation and known as loess. The loess deposit of the Prairie Plains extends westward from a point east of the Mississippi and has its greatest development in Illinois, lowa, Nebraska, and the states to the southeast. The distribution of loess is one of its most important features. It is thick about the borders of the area occu- pied by the Iowan ice sheet, but thins out on interstream areas when traced away from this border tract, though it retains its thickness along the larger valleys.! It follows the Mississippi nearly to the Gulf, and is particularly well developed along this stream and the Missouri. It is habitually found on bluffs immediately overlooking the valleys, and in this position has more than average thickness and coarseness of grain, becoming thinner and finer at a distance from the valley margin.? The northernmost limit of the loess in the central part of the United States is about 35 miles below St. Paul, Minnesota. In central and eastern Illinois and south-central Ohio it passes under the Wisconsin drift. In the Mississippi basin the loess is rarely over a score or two of feet in thickness, though exceptional thicknesses of nearly 100 feet have been noted. It is composed of angular undecomposed particles of calcite, dolomite, feldspar, mica, anda certain number of rarer minerals. It is generally light brown in color and is a variety of silt intermediate in size of grain between the finest sand and clay. Stones are generally absent except at its base. On erosion it often exposes vertical faces for long periods. It is markedly porous, owing in part to vertical tubes usually found in it and supposedly due to root action.? The loess is preponderatingly on the east sides of the main rivers, and this fact suggests the hypothesis of an eolian origin, which has perhaps more to commend it than the rival aqueous hypothesis. It appears that the loess is chiefly wind derived and that favorable opportunities ‘for its formation were afforded by the periodic flooding and drying up of the alluvial deposits along the streams of the region either in the in- 1 Chamberlin and Salisbury, Geology, vol. 3, 1906, p. 407. 2 Frank Leverett, Comparison of North American and European Glacial Deposits, Zeitsch. f. Gletscherkunde, Berlin, vol. 4, 1910, p. 297 ff. 3 For an excellent discussion of the various phases of the loess problem, and for recent data of great consequence, see B. Shimek, Jour. Geol., vol. 4, pp. 929-937, and various loess papers, Bulls. Lab. Nat. Hist., Univ. Iowa, 1904. PRAIRIE PLAINS 489 terglacial period preceding the Iowan glacial invasion or during the Iowan invasion itself. The absence of fluvio-glacial deposits in important amounts on the margin of the Iowan till sheet would appear to favor the idea of a dry glacial stage, strange as this may seem. It is found that the loess is thicker on eastern or lee sides of ridges and prominent hills than on the windward sides, a feature that harmo- nizes with the thicker deposits on the east sides of the stream valleys in pointing to wind action as the chief factor in loess deposition. The diminishing amount of loess in an eastward direction suggests that a considerable part of the material was derived from the Great Plains east of the Rocky Mountains, where dust storms are frequent even to-day. That the loess was deposited in an interglacial period is supported by the fact that buried soil occurs between the loess and the glacial deposits, and appears to indicate a long period between the melting of the ice and the deposition of the loess. This view is strengthened by the finding of a molluscan fauna in the loess strikingly similar to the existing fauna, showing that the conditions which prevailed when the loess was forming were not greatly different from those now existing in the region. The shells found in the loess are almost exclusively those of land species or such as frequent isolated ponds. There is a practical absence of forms that frequent rivers and lakes. Fossils of land mammals, chiefly in the forms of bones and teeth, have also been found. TREE GROWTH OF THE PRAIRIES By definition a prairie is an extensive tract of land, level to gently rolling, without a timber cover. The definition includes the idea of a meadow or at least a grassy vegetation. The distinction between plain and prairie as the terms are actually employed in the United States is not sharply drawn, but in general they are considered to differ in two main respects. The prairie country is dotted with groves of trees and has a continuous grass cover or sod; the plains country is without timber except along the streams, on prominent elevations, and in places about springs; the grassy vegetation of the plains is disposed in clumps sepa- rated by bare spaces. A view in any direction on the typical prairies includes groves and bands of trees and even tracts of true forest. The phrase “Prairie Plains” as here used means the whole great region in which patches of prairie occur and not merely the true prairie land exclusive of wooded tracts (see Plate I and Fig. 23). Since the whole prairie country lies in the belt of moderate to light rainfall between the semi-arid West and the humid East, Fig. 10, it fol- lows that there are great differences in the relative amounts of woodland and true prairie. On the east, north, and south the province borders 1 See especially Proc. Iowan Academy of Science, vol. 15, 1909, pp.57-75.- 499 FOREST PHYSIOGRAPHY heavily forested tracts; on the west it borders an all but treeless region. In general the largest tracts of true prairie within the province are found toward the west in the direction of diminishing rainfall. Locally, however, there are extensive tracts well within the western border. South of Chicago, for example (Fig. | 90), is a large tract known as the Grand Prairie with a rainfall of 35 \| inches a year. Extensive prairie | tracts occur also in northern Indiana, southern Michigan, etc. On the north the Prairie Plains province is covered with the southern edge of the great pine and hemlock forests of the Lake Region. The in- dividual trees attain great size, espe- cially on the sandy glacial material south of the Straits of Mackinac in Michigan; the virgin growth of this district consisted of extensive stands of superb forest. Farther south is the zone of hardwoods with many outliers of the northern pine forests, especially where the land is either too wet or toodry. The sandy, dry situations have a growth of pines; in the wettest situations are growths Fig. 190. — Original distribution of prairie and of cedar and hemlock. In the in- woodland in Illinois, (Barrows, Ill. Geol. termediate situations are woodland Sai edited tom Gone, and a true forest growth of mixed hardwoods. The occurrence of the hardwoods on the richest land is notable and in part explains why they were so extensively destroyed in clearing of land for agriculture. It is a common saying that the soil supporting a good growth of beech and maple forest is better than open prairie soil or land so wet that it must be drained before being cultivated. While the saying has a great deal of truth in it, it must also be noted that the peaty soils of low situations have a high fertility when they are properly drained and sweetened and are better than the upland soils for the production of many vegetable products. What measure of truth the saying contained led the settlers rapidly to clear away the hardwood forests until now the supply of hardwood is greatly diminished and in urgent need of conservation. = Woodland PRAIRIE PLAINS 401 Many owners are beginning to realize the value of their wood lots and are no longer cutting them for fuel but for timber. The strongly marked topographic and drainage features of the heavily glaciated country about the Great Lakes have had an important control . over the forests not only in the distribution of the forest flora but also in the development of the forest products. It is noteworthy that the steeper slopes of the terminal moraines are wooded because too steep for cultivation, just as bottom land and upland are in many places cleared of their original timber cover while the steep undercut river bluffs are allowed to remain tree-covered. The terminal moraines are not only rough, they are also stony and clayey; many miles of terminal moraine are composed of tough stony clay too difficult of cultivation to com- pete with better favored tracts of outwash and ground moraine in the intermorainic belt. Such tracts are commonly left wooded or used for pasture-land or for special crops such as flax and rye. It is also note- worthy that the steep slopes of drumlins in the glaciated country are commonly left in their original wooded condition. Some drumlins have narrow summits, others have broad rounded tops. In the latter case a farm may be located on the summit, in the former case the timber cover may be left undisturbed. In the same way the borders of marshy tracts and the floors of undrained depressions are covered with a growth of swamp vegetation including the swamp oaks and maples, buttonwood, elm, cedar, and other types. The aggregate of such land is large and with suitable improvement might be made to grow a steady supply of valuable timber. On the east the forests of the Prairie Plains grade off into the more or less continuous stands of the Appalachian region in the belt of heavier rainfall. On the south they merge into the pine forests of the Coastal Plain and on the west into the scattered hill and escarpment timber of the Edwards Plateau and its outliers in central Texas. Farther north, as shown in Fig. 23, long finger-like extensions of the Prairie Plains timber follow the stream valleys. For some distance this growth covers both bottom lands and valley slopes and bluffs; farther west it is con- fined to the bottom lands wholly and still farther west it is found only along the margins of the streams. The tree growth of the Prairies of Texas has such a close relation to the geology, water supply, and soils as to warrant a more detailed descrip- tion of its occurrence. The Prairie Plains are here but 100 miles wide and consist of two main subdivisions,— the Black Prairie on the east and the Grand Prairie on the west. The Black Prairie is developed upon a series of marls, sands, and limestones that dip gently east- 492 FOREST PHYSIOGRAPHY ward. The marls have weathered into a uniform and very gently undulating type of topography terminated on the west by a low inward-facing escarpment that never exceeds a relative altitude of 200 feet. It is known as the Black Prairie. West of the Black Prairie is the Grand Prairie, developed upon nearly horizontal limestones of great extent as a series of flat structural plains terminated in succession on the coastal side by low inward-facing escarpments. The western border of the Grand Prairie is a pronounced escarpment with a lobate or crenulated outline that extends from Arkansas, west and south through Oklahoma and Texas, to the Rio Colorado, whence it curves about to the west and north and becomes the eastern escarpment of the Great Plains. The most important forest tracts of this region are the Western and Eastern Cross Timbers, two narrow forest belts developed upon the inner edge of the Grand Prairie and Black Prairie respectively. They are of peculiar interest since they form long narrow ribbons of forest in what is otherwise a prairie country. Both belts have been developed upon the outcrop of two sandy formations (Cretaceous). The open sandy soils not only permit thorough aeration, but also favor the rapid absorption of water during periods of rainfall, whereas the close-textured soils of the intervening prairies shed water to an exceptional degree and are too dry for tree growth. The limited rainfall of the district thus makes the porosity of the soil a determining factor in forest distribution. The Eastern Cross Timbers consist of a belt of timber whose extent is shown in Fig. r9r. It has many outliers towards both the east and the west and like the Western Cross Timbers sends finger-like extensions up and down the stream-ways. Within the Cross Timbers are many local prairie tracts. The Western Cross Timbers are about 10 miles long and have about the same extent north and south as the Eastern Cross Timbers. The various irregularities of the belt in response to geologic changes and to relief are clearly indicated in Fig. tor. The western border of the Western Cross Timbers is the more indistinct since it merges into local forest tracts developed upon other sandy formations. Wherever compact marls and clays outcrop there is a marked absence of forest growth except in cases where the mesquite grows. In general the surface is without tree growth between and on either side of the Cross Timbers; it is here that the prairies of Texas have their most typical development.! Among the effects of glaciation none is perhaps more directly inter- esting to the forester than the overwhelming effects of the continental 1 R. T. Hill, Geography and Geology of the Black and Grand Prairies, Texas, 21st Ann. Rept. U.S. Geol. Surv., pt. 7, 1899-1900, pp. 65-85. PRAIRIE PLAINS 493 Scale of Miles 25 50 LEGEND THE GRAND ‘THE BLACK PRAIRIE REGION PRAIRIE REGION SQ =|] Be WESTERN LAMPASAS FT.WORTH EASTERN BLACK CROSS CUT PLAIN PRAIRIE CROSS PRAIRIE TIMBERS TIMBERS Fig. t91. — Cross Timbers of Texas. ice caps upon the forests whose regions they invaded.!. During the period of maximum development the ice had a most disastrous effect 1 C. H. Merriam, The Geological Distribution of Life in North America with Special Refer- ence to the Mammalia, Proc. Biol. Soc. Washington, vol. 7, 1892, pp. 42 and ff. A094 FOREST PHYSIOGRAPHY upon animals and plants, not only in the great area occupied by it, but far south of its actual border. As the ice advanced from the north, northern species were driven southward and more southerly species correspondingly displaced or exterminated. It appears that the gradual southward movement of the cold zone greatly restricted the range of the plants, compressing them from north to south within very narrow limits. Perhaps the most interesting evidence of the strength of this hypothesis is the presence of colonies of Arctic species on isolated mountain summits in southerly latitudes where at a high altitude abnormally low temper- atures exist similar to those which exist in their northern homes. Such colonies could not in many cases have reached their present position during existing climatic conditions; following the retreat of the ice at the close of the glacial period many boreal species were stranded on mountains where their survival was conditioned by their ability to migrate with a congenial climate. DRIFTLESS AREA Lying well within the southern border of the general region once covered with glacial ice is a small district, Fig. 192, which has never suffered glaciation. It is called the Driftless Area. Its extent is about 10,000 square miles and it lies in the states of Wisconsin, Illinois, Min- nesota, and Iowa, the greater part of it occurring in southwestern Wisconsin. The soil of the Driftless Area is residual, derived directly from the decay of the underlying rock. Like residual soils everywhere the surface detritus grades gradually into solid rock and there is not that sharp dividing line that is found between glacial or other over- placed soils and the rock surface. The surface of the Driftless Area is topographically mature and the drainage is well organized, without falls and rapids and lakes or swamps except within narrow limits and along the river bottoms. The aimless flow of streams developed upon the irregular topography characteristic of till sheets or terminal moraines, as well as the abun- dant swamps, lakes, and ungraded stream courses of such localities, are here practically absent. At numberless points within the area over- ridden by glacial ice the bedrock is scratched, grooved, and smoothed, while within the Driftless Area glacial markings are entirely absent. The thoroughly dissected upland plain of the Driftless Area is in strik- ing contrast to the more even country with drift-filled valleys about it. The valleys and ravines tributary to the main streams have been thoroughly organized and the divides reduced to narrow ridges. The 1 Grant and Burchard, Lancaster-Mineral Point Folio U. S. Geol. Surv. No. 145, 1907, p. I. PRAIRIE PLAINS 495 Co ae ee pee Scale of Miles Fig. 192. — Driftless Area of Wisconsin. valley bottoms toward the south are from 600 to 700 feet above sea level, and the relief of the area is from 200 to 300 feet.’ In spite of the extreme irregularity of the drainage of the glaciated region its relief is less than that of the Driftless Area, a condition due to the manner of distribution of the drift, which in general was lodged in the valleys more freely than on the heights. Such plain-like qualities as the drift-covered region exhibits are due less to the glacial denudation 1J. E. Carman, The Mississippi Valley between Savanna and Davenport, Bull. Ill. State Geol. Surv. No. 13, 1900, p. 29. 496 FOREST PHYSIOGRAPHY of prominences than to the grading up of depressions. In a rough manner the driftless region therefore presents the essential features of the preglacial topography of this portion of the United States.! The purely residual products in the Driftless Area have their greatest expression upon the flat-topped upland areas. On the valley slopes the residual products tend to be removed by erosion, and on the valley bottoms there is alluvial filling derived from the uplands in part during an earlier period of alluviation (probably glacial) and in part by wash under existing conditions.? The Driftless Area has created widespread interest since early in the eighteenth century, largely because it has not that exceptional height which on the border of the glaciated country is the general cause of immunity from continental glaciation. Its average summit level falls a Fig. 193. — Diagrammatic section in the Driftless Area, showing relation of the mantle rock to the solid rock beneath. (Alden, U.S. Geol. Surv.) little short of 1200 feet, while the effective height of the highlands lying between it and Lake Superior is from 1700 to 1800 feet. The summit level between the Driftless Area and the plains of Iowa and southern Minnesota is about 1300 feet, so that as a whole the Driftless Area is somewhat lower than the surrounding tracts. Furthermore, the immunity from glaciation enjoyed by the area was due not to some accidental condition but to a geographically fixed cause, for none of the repeated invasions of the ice affected it.’ The most important elements of the explanation of the Driftless Area are that it is in the lee of the elevated territory of northern Wis- consin and Missouri, which acted as a wedge, forcing the ice away on either side and protecting the region south of it, and that the great valleys of Lake Superior and Lake Michigan lie in such position with reference to the Driftless Area as to tend to divert the glacial streams 1 Chamberlin and Salisbury, The Driftless Area of the Upper Mississippi, 6th Ann. Rept. U.S. Geol. Surv., 1884-85, p. 307. 2S. Weidman, The Geology of North-Central Wisconsin, Bull. Wisc. Geol. and Hist. Surv. No. 16, 1907, p. 554. 3 Chamberlin and Salisbury, The Driftless Area of the Upper Mississippi, 6th Ann. Rept. U.S. Geol. Surv., 1884-85, p. 315. 4 N. H. Winchell, 5th Ann. Rept. Geol. and Nat. Hist. Surv. of Minn., 1877, p. 36. PRAIRIE PLAINS AQ7 to right and left and thus to increase the effect of the highlands lying between the lakes in turning the ice from the driftless region... A third cause is found in the retarded feeble flow and relatively increased wastage caused by the thinning ice through the diversion of the main lobes down the lake troughs. An important factor was the broad zone of glacial wastage which probably includes approximately 150 miles of the ice border; on this assumption the meager ice streams that crept down across the highlands south of Lake Superior and the enfeebled tongues which crept down the adjacent depressions were profoundly controlled by the topography and easily directed this way and that according to the relations of the major topographic depressions. It is also maintained with much reason that the outlines of the ice front would be modified to an important degree by the law of self- perpetuation of climatic conditions; by which is meant that the ice-clad regions increased the snowfall upon themselves and tended toward self- perpetuation, just as the ice-free and therefore relatively warmer area resisted the advance of the ice by excessive melting. While this last agency was not an originating one it would undoubtedly produce a modifying effect and tend to codperate with the topography in holding the surrounding ice to the depressions down which the invasion first took place. The Driftless Area is interesting ecologically because of the occur- rence therein of many plants peculiar to it alone. Among these mosses and other low forms predominate. 1 R. D. Irving, Geology of Wisconsin, vol. 2, 1877, pp. 608, 611, 632, 634. CHAPTER XXV ATLANTIC AND GULF COASTAL PLAIN (INCLUDING THE LOWER ALLUVIAL VALLEY OF THE MISSISSIPPI) GENFRAL FEATURES AND BOUNDARIES THE greater part of the eastern and southern coasts of the United States is bordered by a lowland whose physiographic features are de- veloped upon a mass of soft sands, silts, and clays, chiefly of Cretaceous and Tertiary age, and disposed in strata that incline (dip) gently sea- ward. ‘The structural features are in the main very simple, though in places there are important departures from the average condition. On the north shore of Long Island the clays (Cretaceous) were disturbed by ice action and quite generally concealed by a cover of glacial and fluvio- glacial material; in Louisiana important faults complicate the struc- ture, topography, and drainage; in Oklahoma and Texas some of the coastal plain deposits are indurated, not soft and friable as generally; and in many localities the dip of the strata, almost never great, varies from a mean sufficiently to cause important differences in the width of the coastal tract, the character of the shore line, and the general topo- graphic expression. The Coastal Plain terminates on the north at Cape Cod, on the south in Mexico. Both ends are narrow and the northern is glaciated and partly submerged. The physiographic features of the province are varied at two places by special features of the first order — the peninsula of Florida and the lower alluvial valley of the Mississippi. Its width varies from a fraction of a mile to 500 miles measured normal to the coast; its greatest inland development is in the Mississippi Valley, where it swings northward into western Kentucky, southeastern Missouri, and eastern and southern Arkansas. The inner edge of the northern portion of the Coastal Plain is bordered by water bodies — Cape Cod Bay, Long Island Sound, etc.; by an inner valley lowland in New Jersey, Alabama, and Arkansas; and these features in turn are bordered by various topo- graphic provinces — the Piedmont Plateau developed on crystalline rocks - at the north and east and in succession westward and southward, the Appalachian Mountains and Plateaus, the Ozark Highlands, the Oua- chita Mountains, and the Prairie Plains of central Texas. 498 ATLANTIC AND GULF COASTAL PLAIN 499 FALL LINE The Atlantic division of the Coastal Plain borders the Piedmont Plateau along the fall line, as it is commonly termed. This phrase im- perfectly describes the boundary, which is not really a line but a zone of appreciable width; hence “fall belt” would be a more accurate desig- nation. As the streams cross from the old land or Piedmont Plateau to the new land or Coastal Plain they descend over rapids and falls of moderate height. Between the Raritan and the Roanoke the rivers discharge directly from rock-walled channels into tidal estuaries, since here the depressed land has brought the sea to the very borders of the Piedmont Plateau; but farther south the fall line is above sea level; on the James it is at tide level; on the Neuse 100 feet above the sea; on the Wateree, near Camden, it is 125 feet; on the Savannah, near Au- gusta, 125 feet; on the Chatahoochee, near Columbus, 210 feet; and on the Tuscaloosa at Tuscaloosa, 150 feet. On the interfluves, Coastal Plain and Piedmont Plateau merge into each other in an intermediate zone from a fraction of a mile to 10 or 12 miles wide. The boundary is never a cliff and seldom even a well-defined scarp, and the lowland hills are nearly as rugged and more than half as high as the piedmont hills; but chiefly on account of differences of soil the two provinces present some of the most strongly marked contrasts in the United States, if not in the world. In the Piedmont Plateau the rocks are crystalline, the soils residual, the stream courses flow in narrow gorges with cataracts and rapids; while in the Coastal Plain the soils are derived from clay, sand, and gravel deposits, and the streams flow in shallow valleys and discharge into broad tidal estuaries or coastal swamps. In the piedmont region the topographic details, the soils, and the watercourses reflect the character of the rock, while in the coastal region these features represent the work of streams born upon plains newly uplifted from the sea, or reflect the general attitude of the lowland rather than its rock character. Among the important cities located upon the common boundary of these two great provinces may be mentioned Raleigh, Camden, Columbia, Augusta, Macon, Colum- bus, Montgomery, Tuscaloosa, Little Rock, Austin, San Antonio, etc. RELATION TO THE CONTINENTAL SHELF The seaward limit of the coastal lowlands on the southeastern border of the United States is a steep scarp which marks the transition from the outer edge of the shallow continental shelf to the abysmal depths of the main ocean floor. This scarp forms the true continental margin and is 500 FOREST PHYSIOGRAPHY comparable in height and extent though not in topographic variety with some of the most majestic mountain ranges. Were sea level lowered to its foot, present sea level in this latitude would be a region of ice and snow. The seaward slope of the Coastal Plain is-continued as the continental shelf beneath the level of the sea to distances varying from too to 200 or more miles; the Coastal Plain is therefore to be considered as the landward extension of a greater plain whose outer border is submerged. The present position of the shore line upon this plain is a purely accidental one and subject to change, indeed is now changing at what from even a human standpoint may be considered a rapid rate. The steady depression of the larger part of the coast is so pronounced that if continued the sea will in a very short time again cover the coastal lowlands. In more recent periods of the earth’s his- tory such transgressions of the sea over the land have been frequent and important, and no less important have been the various retreats of the sea which have brought the coastal border of the continent out near the border of the continental shelf and turned what is now conti- nental shelf into dry land. The repeated invasions of the sea are clearly indicated by the marine fossils found in the flat-lying sediments which constitute the Coastal Plain; and the surface of the continental shelf is deeply scored opposite the mouths of many of the coastal rivers by trenches or sub- marine canyons whose position and character have led to the conclusion that they represent old river valleys formed at a time when the land stood higher than now. MATERIALS OF THE COASTAL PLAIN ATLANTIC SECTION The lowest strata of the Atlantic Coastal Plain, as illustrated in the Maryland section, are composed of arkosic sands and clays derived from a deep mantle of disintegrated gneiss and phyllite such as now form part of the Piedmont Plateau. The sands and clays have de- tached outliers west of the main border of the Coastal Plain which show by their position, altitude, and composition that the plains strata formerly extended farther west over the adjoining portion of the Pied- mont Plateau. Higher in the section the strata consist of variegated clays and coarse, irregularly bedded sands. Then follow sands and clays in alternation with but slight variations in character and a gradual transition toward sandy and marly deposits.! 1 C. Abbe, Jr., A General Report on the Physiography of Maryland, Including the Develop- ment of the Streams of the Piedmont Plateau, Md. Weather Serv., vol. 1, pt. 2, pp. 74-75. ATLANTIC AND GULF COASTAL PLAIN 501 GULF SECTION The Gulf Coastal Plain section is rather typically represented in the Arkansas-Louisiana district where the lowermost beds are sandy and contain vegetable remains and brackish-water shells. Above these basal beds are limestones and marls; higher in the section are sands, lignitic clays, marls, and chalks, a succession indicating a gradual deepening of the water in which the sediments accumulated. Succeeding the deep- water deposits are limestones, marls, and lignitic sands, showing a return to shallow-water conditions.'. Then came complete and final uplift of the region and the formation of a coastal plain out of what had long been a continental shelf. The migration of the shore zone, characterized by sand deposits, over the surface of the plain from its inner to its outer edge was the last important geologic event and is the cause. of the prevailingly sandy nature of the surface deposits. SUBDIVISIONS (t) Including a broken and glaciated northern section the Coastal Plain may be said to fall into seven well-defined districts. The first includes Cape Cod, Martha’s Vineyard and Nantucket islands, Long Island, and a number of lesser islands. (2) The second extends from the Hudson to the Potomac, and is char- acterized by a cuesta, an inner lowland, and a gentle outward slope or lowland merging into great estuaries that nearly isolate it from the mainland. (3) The third extends from the Potomac to the Neuse. It is a low east- ward-sloping plain characterized by long arms of the sea reaching far into it, by broad terraced plains of loam, and broad coastal sounds derended cn the seaward border by long tenuous reefs partly of sand, partly the narrow and wave- and current-modified outcrops of clayey formations (Cretaceous) that dip seaward.’ (4) The fourth section extends from the Neuse to the Suwanee as a gently sloping plain characterized by great stretches of pine-covered sands with swampy marginal development also fringed by coastal reefs. (5) The fifth section extends from the Suwanee River to the bluffs at the eastern border of the Mississippi flood plain. This division has more important topographic irregularities than the Georgia-South Carolina section of the Coastal Plain, its inner border having been dissected in 1 A. C. Veatch, Geology and Underground Water Resources of Northern Louisiana and Southern Arkansas, Prof. Paper U. S. Geol. Surv. No. 46, 1906, pp. 20-28. 2 Collier Cobb, Notes on the Geology of Currituck Banks, Jour. Mitchell Soc., vol. 22, No.1, Daw7e 502 FOREST PHYSIOGRAPHY such manner as to develop a rather typical inner lowland and cuesta. Along the coast are long narrow keys, some of which are scarcely able to maintain themselves owing to the depression of the land and to the encroachment of the waves, while some are completely submerged and have the form of offshore shoals. The seaward border of the Coastal Plain of this section is marked by broad lagoons, low coastal swamps, and extensive savannas. (6) The sixth section is the low, poorly drained, swampy terminus of the flood plain of the Mississippi, an area whose southern end is scarcely yet reclaimed from the sea. (7) The seventh section extends from Atchafalaya Bayou to the Rio Grande. Its interior development is topographically weak except in Arkansas and Louisiana, where a somewhat regular series of cuestas and inner and outer lowlands has been developed. The seaward margin of this district is bordered by swamps, extensive sounds, and long narrow lagoons.1 CapPpE Cop-Lone ISLAND SECTION CapE Cop The northernmost section of the Coastal Plain is broken by dissection and drowning into a number of islands, shoals, and capes of which the most important members are Cape Cod and Long Island. Lesser members are Staten Island, Block Island, and Martha’s Vineyard and Nantucket islands. : That these are all members of a formerly more extensive and less dissected coastal plain fringe is inferred from (1) the evidences that stream courses on the present mainland border are superposed from a former cover of gently and regularly sloping marine deposits; (2) the actual finding of fragments of such deposits at Marshfield,? Boston,’ and Third Cliff,4 at the first two localities below sea level and at the third above it; (3) the occurrence of known Tertiary de- posits® on Georges Bank east of Cape Cod in line with the general trend of the strike of the unsubmerged portions of the plain to-day; and (4) from the general likeness of the shoal and channel outlines of the sea floor north of Cape Cod to drowned valleys.6 1 For a characterization of these various districts and a clear description of the coastal plain of the southeastern portion of the United States, see W J McGee, The Lafayette Formation, r2th Ann. Rept. U. S. Geol. Surv., 1890-91, pt. 1, pp. 353-521, with four excellent maps, photographs, and cross sections. 2 C. H. Hitchcock, Final Report on the Geology of Massachusetts, vol. 1, 1841. * F. G. Clapp, Clay of Probable Cretaceous Age at Boston, Massachusetts, Am. Jour. Sci., vol. 23, 1907, pp. 183-186. 4 I. Bowman, Northward Extension of the Atlantic Preglacial Deposits, Am. Jour. Sci., vol. 22, 1906, pp. 313-325. 5 A. E. Verrill, Occurrence of Fossiliferous Tertiary Rocks on the Grand Bank and Georges Bank, Am. Jour. Sci., 3d Series, vol. 16, 1878, pp. 323-332. 6 N.S. Shaler, The Geology of the Cape Cod District, 18th Ann. Rept. U. S. Geol. Surv., pt. 2, 1896-97, pp. 516, 578, 580. ATLANTIC AND GULF COASTAL PLAIN 503 Cape Cod is the most extreme projection on the eastern coast of the United States north of the peninsula of Florida. In general outline it roughly resembles a great flexed arm, the southern portion being the humerus, the eastern portion the forearm, and the northernmost ex- tremity the clenched fist. It is so narrow at the base where it adjoins the mainland that a canal is building to connect Cape Cod Bay and Buzzards Bay; the land section will be only 8 miles long. Its exposed position, the lack of good harbors, and its featureless shores combine to give the Cape notoriety as one of the two graveyards of the Atlantic. The projection of the Cape eastward and northward would not seem so remarkable if the material composing it were solid rock of great resistance to wave erosion; as a matter of fact no hard rock occurs upon it, only soft, easily washed gravel, sand, and clay. The extreme northern end of Cape Cod, the wrist and fist of the Cape Cod arm, is composed of wave- and current-derived material entirely. It is low, covered with sand dunes of irregular outline, bears a scanty growth of shrubs and grasses or is entirely bare of vegetation, and has well-rounded coastal outlines where the wear of currents and waves is constantly reshaping the beach. The material added to the Cape at this point is derived from the crumbling eastern side of the arm. The attack of the waves on the steep eastern shore is vigorous and sustained, and at the present rate of retreat, 3.2 feet per year, the Cape will be entirely destroyed 8000 to 10,000 years hence. On the western side of the Cape the shore is low and flat and bordered by wave-built sand reefs which enclose extensive lagoons. Cape Cod is relatively quiet, a great natural harbor, yet the shore currents are here attacking the border of the Cape and assisting in its demolition.!. The same action is exhibited at Monomoy Point, the southeastern extremity of the Cape, where a portion of the material torn from the bold eastern margin of the Cape is drifted southward and built into a long flying sand reef. The foundation of a large part of Cape Cod is a mass of preglacial sands and clays pre- sumably of Cretaceous or Tertiary age, revealed in well sections, clay pits, cuts, and bluffs. These deposits form the substructure of the east-west section of the Cape where they occur above sea level. The north-south section nowhere reveals similar deposits above sea level, but they have been encountered in well borings. The morainal ridge that forms so promi- nent a feature of the first-named section rests upon these preglacial deposits and its height is enhanced thereby.? The surface material of the Cape is, however, not derived from the basement deposits; it is of glacial derivation. It occurs in the form 1 W. M. Davis, The Outline of Cape Cod, Proc. Am. Acad. Arts and Sci., vol. 31, 1806, Pp. 331-332- 2 N.S. Shaler, The Geology of the Cape Cod District, 18th Ann. Rept. U. S. Geol. Surv., pt. 2, 1896-97, D. 534. 504 FOREST PHYSIOGRAPHY of a thin veneer of till except (1) where actual morainic ridges of excep- tional thickness occur, or (2) where outwash deposits were laid down in front of the moraine. Cape Cod is therefore a mass of glacially derived material resting upon a preglacial basement of sands and clays. Its detailed surface features are due chiefly, and in some cases wholly, to ice action. Lakes are relatively abundant; the surface is character- istically pitted with kettle holes; unsorted material, till, is common; and there are many erratic ice-rafted bowlders. The glacial material is prevailingly sandy and this also is the character of the outwash plains that front the moraine. Lying well forward of the main line of the coast where no topographic prominences afford shelter from the wind the Cape is exposed to the fury of every tempest and its porous, light, dry material has been largely blown into dunes and drifts of sand. This is especially true on the immediate shore, where dunes of’ considerable height occur and where serious effort has been made to stop the progress of the wind-drifted material. CONTROL OF SAND DUNES The problem of dune control is a paramount one in the utilization of large portions of the surface of Cape Cod. Similar efforts are required in many other localities in the United States, and since these are more or less alike in natural features and involve the same problems, the fol- lowing general discussion is included here. The chief areas of shifting dunes to be found along the Atlantic coast are on Cape Cod, near Provincetown; southern New Jersey, near Avalon and Stone Harbor; Cape Henlopen, near Lewes, Delaware; Cape Henry, Georgia; Currituck Banks, North Carolina; Isle of Palms, near Charleston, South Carolina; and Tybee Island, near Savannah, Georgia. Sand dunes are also found on the Pacific coast in Ventura, Monterey, and Mendocino counties, California, and along the coast of Oregon.! In California drifting sand has played an important part in obstruct- ing the channels in certain harbors on the coast; for example it has been blown into the channels every year near Eureka from a narrow sand spit several miles long between Humboldt Bay and the ocean.? Very 1 A. S. Hitchcock, Controlling Sand Dunes in the United States and Europe, Nat. Geog. Mag., vol. 15, 1904, Pp. 43-47. 2 S. B. Kellogg, Problem of the Dunes, Cal. Jour. Tech., vol. 3, 1904, pp. 156-159. See this paper for a discussion of various kinds of schemes for controlling sand dunes by sand- binding grasses, such as beach grass (Ammophila arenaria) and ‘‘rancheria grass” (Elymus arenarius); and for a photograph of a dune covering a forest above Fort Bragg, Mendocino County, Cal. ATLANTIC AND GULF COASTAL PLAIN 505 troublesome dunes are developed on a large scale along the Columbia River in Oregon and Washington from Dallas to Riparia, from sand blown about over the flood plain of the Columbia, which is widely ex- posed during the dry season. Dunes occur in large numbers also about the southern end of Lake Michigan, on the eastern side of Lake Michigan, and east of Lake Huron. The most important principle of control of sand dunes is the establish- ment of a cover of material that will prevent drifting, the type of cover usually depending upon climatic conditions, cost of material, etc. Where possible it is best to produce a forest cover, for this is permanent and, if properly managed, yields anincome. For this purpose the land must be temporarily fixed in some other manner than by seedlings, and a tem- porary binding of inert material such as brush, sand sedges, etc., may suffice. Beach grass has given most satisfactory results in the Great Lake region, North Carolina, and Europe. The grass is transplanted in the spring or fall and set 2 or 3 feet apart in the sand. It has the power of continuing to grow up through a cover of drifted sand by establishing root systems at constantly higher levels. If the sand is temporarily fixed by these or other means, young trees, usually conifers, may be planted and a forest established. In southwestern France a. forest has been established by sowing the seed of Pinus maritima upon the sand and covering with brush. In France and on the coast of Prussia it has been found necessary in places to construct long artificial barrier dunes between-the ocean and the forest and to fix the barrier dunes by beach grass and maintain them by constant oversight. In northern Europe the trees employed for reclamation of sand dune tracts are Pinus montana, Pinus laricio, Pinus austriaca, and Pinus sylvestris.! Extensive pitch-pine plantations are now being experimented with on the sandy areas of Cape Cod, Martha’s Vineyard, and Nantucket. NANTUCKET AND MARTHA’S VINEYARD The relief and drainage of Nantucket Island afford many parallels to the conditions on both Cape Cod and Long Island. There is a terminal moraine on the northern shore of the island and a very irregular pri- mary or inner shore; the same kind of outwash plain is built forward of the moraine, pitted with kettle holes and creased by shallow drainage ways dry for the greater part of the year; and long tenuous sand reefs and flying spits occur like those on the coasts of the adjacent islands. Underneath the thin mantle of glacial materials lie preglacial deposits 1 S. B. Kellogg, Problem of the Dunes, Cal. Jour. Tech. vol. 3, 1904, p. 47. 2 Yearbook Dept. Agri., 1909, p. 336. 506 FOREST PHYSIOGRAPHY which in places project above the level of the sea, so that were the glacial accumulations removed a number of small islands would still exist where Nantucket stands. For the most part the older surface is masked by glacial till and outwash gravels. The original outline of the island has been much altered by wave and current action; portions of the shore have been cut away, while other portions have been fringed with recently formed marine deposits.! INDIAN HILL Hae PAULS POINT ogee ee SEA LEVEL §S.E. KAMES TERTIARY 2 FRONTAL MORALNE KAMES TERRACE DRIFT Fig. 194. — Diagrammatic section of Martha’s Vineyard. (Shaler, U. S. Geol. Surv.) Martha’s Vineyard lies west of Nantucket and has closely allied physiographic features. Not only are the deposits of the island of the same general nature but they are also disposed in roughly the same way, so that the northeastern and northwestern sides of the triangular island are hilly and morainic, with irregular shores, while the southern side is an almost straight eastward-trending line. It presents almost wholly those features typical of a morainal island. The surface drift is almost everywhere thin (about 1o feet) and as on adjacent islands it to some extent masks the older basement material, Fig. 194. The straight southern side of the island is a sand reef enclosing a large number of water bodies with a branching pattern closely resembling that of a stream system and tributaries, a condition signifying a slight sub- mergence of the land after the formation of the outwash plain bordering the shore.? Lone IsLaAnD GENERAL GEOGRAPHY To a forester Long Island is of peculiar interest. It supports a con- siderable variety of forest trees under special conditions, and the rela- tions of these to the physical conditions are the more easily determinable because of the detailed studies made by the United States Geological Survey and the Soil Survey. The value of the natural resources of Long Island is unusually great because of their proximity to a great consuming center; and detailed studies of its climate, soil, water supply, and physiography have therefore been made. The greatest length of Long Island (the ‘“‘Lange Eyelandt ” of the early Dutch) is 118 miles and its greatest width 23 miles. Its outline suggests the shape of a huge fish; the flukes 1 Curtis and Woodworth, Nantucket, a Morainal Island, Jour. Geol., vol. 7, 1899, pp. 226- 236; N.S. Shaler, The Geology of Nantucket, Bull. U. S. Geol. Surv. No. 53, 1800. 2 For geologic description see N. S. Shaler, Report on the Geology of Martha’s Vineyard, 4th Ann. Rept. U. S. Geol. Surv., 1885-86, pp. 303-363. ATLANTIC AND GULF COASTAL PLAIN 507 of the tail are represented by Orient and Montauk points on the east; the head is represented by the blunt western end with its mouth-like extension at Coney Island. The south shore is double: the inner or primary shore line is the border of a broad lagoon — Great South Bay and its extensions; the outer shore line consists of narrow sand reefs of remarkably regular outline. The eastern half of the northern shore is without notable indentations, but the west- ern half is deeply embayed by a dozen or more well-developed fiords with steep sides and note- worthy depth of water. ‘GLACIAL TOPOGRAPHY TERMINAL MORAINES The principal relief features of Long Island are associated with a double line of eastward-trending glacial moraines fronted by extensive outwash plains of loose alluvium, Fig.195. The southernmost is known as MOCO SI S ey SSA ; ha} 4 OPO xy ‘OB g MARTHAS “eags ~ © VINEYARD D, _ ‘LEGEND VA7) Covered with icé at the Ronkonkoma stage [ANY Covered with ice at the Harbor Hill stage ‘ }Sandy Hook Fig. 195. — Relative positions of the ice during the two stages of the Wisconsin glaciation. (Veatch, U. S. Geol. Surv.) the Ronkonkama moraine. Its eastern extension gives character to that end of Long Island, and with the eastern extension of the Harbor Hill, or northern, moraine encloses a large body of water known as Peconic Bay. The Ronkonkama moraine is remarkable for the large body of water it encloses and after which it was named, Lake Ronkonkama. This is by far the largest lake on Long Island and extends about 25 feet below sea level. The two moraines are often referred to as the back- bone of Long Island, though it should be noted that it is a compound backbone, for east of Huntington the two moraines diverge more and more and enclose broad sub-level tracts between them. The seaward slope of a large part of the eastern half of the northern moraine has been largely removed by wave and current erosion, and in a few localities this has been carried so far that in the wave-cut cliffs are exposed sections of the outwash plain through the complete removal of the moraine once fronting and guarding it on the north. Everywhere along this shore the projecting headlands are being modified by vigorous wave action. Cliffs are kept 508 FOREST PHYSIOGRAPHY perpetually freshened and so steep that landslides are a common occurrence, bars are being built across the bay mouths, and in the mill of the surf the bowlders and stones of the crum- bling moraine are fast being ground to pieces. The Montauk moraine lies chiefly in the interior of the island and is not subjected to such general attack by shore processes though its eastern extension, exposed to all the fury of Atlantic storms, is being battered rapidly to pieces. That the extreme eastern end must once have extended farther east is self-evident. Erosion has now progressed so far that portions of the once more extensive outwash plain have been destroyed. Both moraines bear the usual marks of terminal accumulations formed by a continental ice sheet. Their surfaces are deeply pitted by large and small kettle holes, and enclosed, formless pits and depressions, and a maze of mounds, knobs, and ridges of till and other glacial débris still further diversifies their surfaces. Upon the floors of some of the en- closed depressions lakelets or swamps occur, and ‘all others are dis- Fig. 196. — Section showing the relation of outwash to terminal moraine. (U.S. Geol. Surv.) tinctly moist, but by far the larger number are without standing water. Large portions of both moraines are composed of till of rather typical composition, but there are also large portions which are composed largely and in some localities almost exclusively of sand. The sandy phases are developed chiefly at the eastern end, the clayey phases at the western end of the island. Bowlders, large and small, are scattered freely upon the surfaces and throughout the mass of both moraines. They are of variable composition; gneisses and _ schists, sandstones and quartzites predominate. OUTWASH PLAINS Long gently sloping outwash plains extend southward from both moraines. Their northern margins are in many places pitted by kettle holes, and a considerable number of old broad drainage grooves cross them from north to south. The grooves are not now effectively occu- pied by streams, for the present drainage consists chiefly of small wet- weather streams extremely diminutive as compared with the ancestral streams whose channels they occupy. The old broad channels have a markedly asymmetric development, the steepened slope being on the west, a response, it is thought, to the right-handed deflective influence of the earth’s rotation in the northern hemisphere.! The porous nature of the outwash material greatly reduces the run- 1G. K. Gilbert, The Sufficiency of Terrestrial Rotation for the Deflection of Streams, Am. Jour. Sci., 3d Series, vol. 27, 1884, pp. 427-432. ATLANTIC AND GULF COASTAL PLAIN 5090 off, and instead of a normal run-off of about 40% the run-off is about 60% of the total precipitation. The reduced run-off has retarded the dissection of the frontal plains, and the small elevation has favored the Fig. 197. — Outwash plain in foreground, the hills in the background are preglacial and have but a slight morainal covering. (Veatch, U.S. Geol. Surv.) same result. They extend mile after mile as almost flat plains of allu- vium in striking contrast to the ruggedness and picturesqueness of the terminal moraines. = & me) 3S me) errs) 2 o 2 = 2 O f= sy a= c 3 pete o cx 2 Cc D 5 o © Long 2 s = Ute 2 Island a a SS OFT OUT. Lead) ee cae NN faa} A DIGS SII NNN a ————— az. y ————— A BirAor a Gnelss a) 7 Cea Cretaceous— : i Scales 3 Horizontal Vertical 543210 5 10 miles 1000500 0 1000 2000 ~=3000 feet SS ——————————— Tr Fig. 198. — Cross section of Long Island, showing glacial deposits (dotted) and pre-Cretaceous peneplain (Cr.). (Veatch, U. S. Geol. Surv.) TOPOGRAPHIC FEATURES RELATED TO STRUCTURE The present topography of Long Island can not be fully understood without some knowledge of the substructure, Fig. r98. Cretaceous and Tertiary sands of great thickness occur up to heights of several hundred 510 FOREST PHYSIOGRAPHY feet above sea level; their outcrops are exposed principally along the north shore, where the long, deep, and narrow fiords or reéntrants re- veal good sections of the white and red sands and white clays of the Cretaceous formations. In the Half Hollow and Mannetto hills and distinctly beyond the limit of the ancient ice sheet preglacial sections also are exposed, and the strata (Tertiary) consist of fine fluffy sands without glacial material of any sort. From such exposures of pre- glacial material above sea level it follows that an island would now exist even if the glacial material were removed; and in fact such an island did exist before the glacial period, but it was an island of far gentler topography and much more regular outline. Upon a base-leveled and down-warped rock floor Cretaceous and Tertiary materials were laid down and afterward eroded as the result of uplift of the sea floor. The erosion was at first normal and resulted in the formation of an inner lowland where now Long Island Sound occurs. Fol- lowing this there were a number of emergences and submergences of complex character, during which the drainage of the inner lowland was first developed to the west and later, through the tilting of the land, to the east. The essential features of the present topography were developed after the Lafayette sub- mergence. The erosion of the inner lowland was continued to the point where a well-developed cuesta was formed, and it is to the further accumulation of glacial material upon the summit of this cuesta that Long Island owes its present marked contrast between rugged moraine and abrupt high northern shores on the one hand and smooth southern plain with low, reedy, flat southern shores on the other. The scouring action of the ice sheet further deepened the inner lowland, and with the final disappearance of the ice from the region a slight sinking of the land brought out even more strongly the features owing to depression. At a higher elevation the inner lowland of Long Island would have similar relations to the cuesta and the old-land as the inner lowland of New Jersey now has to the cuesta of the New Jersey coastal plain on the east and the old-land on the west. SOILS AND VEGETATION SOIL TYPES A soil map of Long Island, Fig. t99, shows four broadly defined types of soils: (1) stony loams and gravels which occupy the terminal moraines and the narrow plateau between the northernmost moraine and the north shore, (2) coarse sandy loams that constitute the greater part of the outwash plain, (3) fine sandy loams which form the outer fringe of the outwash plain and those portions of it that are adjacent to the old drainage ways, and (4) clay loams that form a transition type between the upland and the salt marsh and the beach sands. While the clay loams have greater natural fertility than the sandy loams they are often found on lands too rough for cultivation, and the great market gardens of the island are found on the outwash principally, where many of the conditions of cultivation are ideal. Agriculture has become so highly specialized in the western part of the island on account of prox- imity to New York City that the natural fertility of the soil is a far smaller factor than position with reference to the market. ATLANTIC AND GULF COASTAL PLAIN Lp ia 72° Aa oe Frum IB Tf SX GARDINERS 1, NEW YORK v. =e. vee De da Ronkonkoma i Na: \ LC MAP OF LONG ISLAND, NEW YORK 5 ° Beant On0eed 5 SCALE IN MILES @@ TERMINAL MORAINES IN BLACK GARDINERS 1, SHELTER TF ye Montauk Pts NEW YORK 2 y < K 41 5 : oot a Pe MZ ag Harbor _ Smithtown Fo ee Freat ue: ¢g MAP OF LONG ISLAND, NEW YORK Sl atO Set On M15 SCALE IN MILES SOILS STONY LOAMS AND GRAVEL GRAVEL AND COARSE SANDY LOAMS SAND (NORFOLK) CLAY LOAMS, FINE SANDY LOAMS, BEACH SAND, SALT MARSH NEW YORK 41° MAP OF LONG ISLAND, NEW YORK SG) Swe Ovarts SCALE IN MILES FOREST TYPES PRAIRIE, SCRUB OAK, STUNTED OAK PITCH PINE ETC. MIXED HARDWOODS ETC. Fig. 109. —- Terminal moraines, soils, and vegetation of Long Island. Se heck em Fig. 200. — Characteristic growth of pitch pine (background) and scrub oak (foreground) on the outwash plains of eastern Long Island. Fig. 201.— The effects of repeated fires on soil and vegetation, Long Island. The raw soil with the humus almost entirely burnt out has the appearance of snow in the photograph. 512 * ta a i B t\ H LH st | Fig. 202. —Typical growth of hardwood on the clayey portions of the Harbor Hill moraine, Long Tsland. Fig. 203.—Scattered growth of pitch pine and scrub oak on the sandy portion of the Ronkonkama moraine south of Riverhead, Long Island. 513 514 FOREST PHYSIOGRAPHY NATURAL VEGETATION In the disposition of the natural vegetation we have a far truer index of soil fertility and water-supply conditions than in the artificial dis- position of farms. On the poor, dry, sandy, porous soil of the outwash plains the characteristic growth is pitch pine and scrub oak (Quercus nana and Quercus prinoides), as Shown on the accompanying map, Fig. 200. The natural prairie for which Long Island is famous is also located on the southern outwash plain in the vicinity of Hempstead and Garden City. It bears a growth of prairie grass and has never been covered with forests. In Dutch colonial days it was a famous pasture ground for sheep, horses, and cattle. On the better soils of the outwash plains a better growth of pitch pine is found, and even a small stand of white pine has been reported southeast of Sag Harbor. Where a slight admixture of clay or loam appears, as in a number of scattered loéalities, a few species of hardwoods grow with pine and oak. Upon the moister and more clayey moraines excellent growths of hard- woods occur, Fig. 202. Chestnut, oak, elm, beech, and locust constitute the principal types. In a few localities, as several miles southwest of Port Jefferson, the moraine and accompanying outwash are occupied by oak, but it is a stunted growth and reflects the infertile character of the Norfolk sand on which it grows, the distribution of the stunted species corresponding almost precisely with that of the Norfolk sand. The sandy phase of the Ronkonkama moraine south of Riverhead is occupied by pitch pine, which occurs up to the summit of the moraine, Fig. 203. UMM VUES ee )) Fig. 207. — Albemarle and Pamlico Sounds and bor- Fig. 208. — Chesapeake Bay and Delaware Bay and dering sand reefs, east coast of North Carolina. the principal bays tributary to them. SouTH CAROLINA-GEORGIA SECTION | That portion of the Coastal Plain between the Neuse River of North Carolina and the Suwanee of southern Georgia and Florida is a land of gentle slopes which incline from the fall line to the coast. It is a pine- clad, sandy plain with dendritic drainage. The seaward margin of this section is a wave- and current-built sand reef at the north and a line of sea islands at the south, that merge at the extreme south into long low islands locally known as “keys.” Along the Altamaha River the overflowed river bottoms have been re- claimed to some extent by diking and ditching and are cultivated. Their position is excellent for rice culture, since irrigation water is easily applied. Agricultural operations have been confined principally to the diked river lands since the early part of the nineteenth century, and rela- tively small areas have been cropped on the uplands since the early 520 FOREST PHYSIOGRAPHY abandonment of indigo growing and the decline and cessation of cotton production.!. Near Savannah these diked rice fields have been badly neglected in recent years, but the fertility of the tide-marsh soils is beyond question, and were drainage reéstablished they weuld consti- tute a valuable addition to the farming lands of the coast.? In portions of eastern North Carolina the seaward margin of the Coastal Plain bears drainage ways which are merely natural depressions marked by a quick growth of water-loving shrubbery and in some areas no well-developed drainage courses occur. These are the savannas, or open flat lands with badly drained soils that support a poor growth of pine and an undergrowth of berry bushes and bay and pitcher plants.® Inland from the low coastal zone is the broad belt of pine-clad sands which more than anything else characterizes the section. It is a vast plain with slight undulations, the depressions slightly accented by streams occurring in the form of old terraced scarps much dissected owing chiefly to the friable character of the material, and now having rounded bottoms and softly contoured sides. With increasing distance from the sea the land stands higher and higher, the streams are more active, the val- leys deeper, and the surface more undulating, often rising into rounded hills. Again, as in the north, the hills may be isolated and flanked by terraces or may constitute parts of broad interstream areas, and some- times the terraces extend a short distance into the Piedmont Plateau. The common boundary of plateau and plain is ill-defined as to topog- raphy but rather sharply marked as to drainage and soil characters. ALABAMA-MISSISSIPPI SECTION The great section of the Coastal Plain between the Suwanee and the Mississippi is topographically diversified by two pronounced scarps; the first one is the steep river bluffs along the eastern border of the Mis- sissippi flood plain, bluffs due to stream planation; the second is the cuesta of the Alabama-Mississippi section of the plain. The cuesta begins on the river bluffs overlooking the Mississippi in extreme western Kentucky, crosses the Tennessee-Mississippi boundary 50 miles east of the river bluffs, curves southeastward to within 50 miles of the head of Mobile Bay and dies out eastward. This feature represents the inward- facing slope of the dissected Coastal Plain of Alabama and Mississippi. It 1 Milton Whitney, Soils in the Vicinity of Brunswick, Georgia, Cir. U. S. Bur. Soils No. 20, i anton Whitney, Soils in the Vicinity of Savannah, Georgia, Cir. U. S. Bur. Soils No. 1), D./-2. 3 Milton Whitney, Soils of Pender County, North Carolina, Cir. U. S. Bur. Soils No. 20, PD: I=2. ATLANTIC AND GULF COASTAL PLAIN 521 stands 600 or 700 feet above sea level. The outer lowland averages 200 feet lower, while the inner lowland is a broad trough often roo and some- times 200 feet lower than the border of the cuesta. The inner lowland is from 20 to 25 miles on the average and extends eastward a short distance beyond Montgomery. The densest population of the state of Alabama outside the cities is found in the inner lowland where the Selma chalk has weathered into a tract known as the “black belt”? because of its prevailingly black soils. These are highly calcareous residual clays and have a black color where they contain much organic matter. They are among the most fertile lands in the South. In the eastern counties of Alabama a limestone (Gladden) of the coastal-plain series, 200 feet or more thick, is extensively developed into caves and lime sinks with which are associated big springs. This for- mation is marked by the occurrence of strong limy black soils similar to the black prairie soils of the Selma chalk, but the topography is so broken and deeply eroded as to be in sharp contrast to the smooth-floored Selma lowland. After the Tertiary period a blanket formation was deposited upon the Coastal Plain. This is known as the Lafayette formation and is a man- tle of reddish and light-colored loams and sands with frequent beds of water-worn pebbles in the lower parts. It is from 25 to 30 feet thick on the average and formerly covered the entire Coastal Plain, resting un- conformably on the older formations. In general it is sympathetic to the topography, though in many large areas it has been in great part removed by erosion. Because of its widespread development it consti- tutes about four-fifths of the cultivated soil of the entire Coastal Plain of Alabama, and is the chief factor in affecting the character of the soils.! The Cretaceous and Tertiary beds of the Coastal Plain of Alabama have an average dip toward the Mississippi and the Gulf of Mexico of 30 to 40 feet a mile. The surface of the Coastal Plain descends in the same direction at a much less rapid rate, about a foot per mile, so that in going toward the south from the Appalachian Plateaus one passes in succession over the beveled edges of these deposits from the oldest to the newest. Each formation with few exceptions occupies the surface in a belt proportional in width to its thickness and running nearly east and west. The sandy formations of the outer lowland are commonly character- ized by short steep slopes and frequent ravines, the shales by long 1 E. A. Smith, The Underground Water Resources of Alabama, Geol. Surv. of Alabama, 1907, p. 12. 522 FOREST PHYSIOGRAPHY slopes and few ravines, and the calcareous formations by smooth-con- toured ill-drained valleys known as ‘black prairies.” The inland margin of this district is even less definite than the neighboring region to the east. In western Georgia and eastern Alabama the rivers cascade over hard rocks to form sluggish stretches in the lowlands, but commonly an arm of sedimentary material extends miles into the adjacent plateau in an ancient estuary and the river transition is seldom sharp. In central Alabama, where the Coastal Plain overlaps the southern end of the Appalachians, long fingers of lowlands stretch into the valleys between the ridges. In northwestern Alabama the line of demarcation between the Cumberland Plateau and the Coastal Plain is so poorly defined that it can not be drawn except as a zone from to to 20 miles wide. Still farther north the boundary between the Coastal Plain and the older formations coincides approximately with the course of the Tennessee River, though here and there occur outcrops of the older harder rock west of the river and outcrops of softer coastal-plain deposits east of the river. The trunk streams of the Alabama section of the Coastal Plain flow across the Cretaceous and Tertiary strata, while the tributaries flow in general parallel to the strike of the outcrops. The infacing or north- ward-facing slopes of the hills are precipitous, while the southward- facing slopes are gentle. The tributary streams generally flow at the base of the steep infacing slopes. The major streams of the region thus run transversely to the cuesta and preserve their ancient consequent courses gained after the last emergence of the coastal lowlands from the sea. The tributary valleys on the other hand respond to a large degree to the detailed geologic structure and have excavated abnor- mally large subsequent valleys along belts of weaker strata. They thus join the master streams almost at right angles, and their direction of flow conforms to a notable degree to the outcrop of the strata upon which they have been developed. The outer edge of the Coastal Plain of Alabama and Mississippi is rather sharply defined by a line of more or less dissected bluffs and hence appears as an upland when viewed from the south. The un- dissected interfluves are flat everywhere except for shallow depressions (of uncertain origin) sometimes containing ponds and bordered by a shrubby growth of gums. The border depressions are without stand- ing water and are usually grassy savannas or pine meadows without undergrowth.! The southern border of this section of the Coastal Plain is in part not a coastal plain but a flood plain under the dominance of the Mis- sissippi River. It is commonly skirted by keys separated from the mainland by narrow sounds, but the keys are narrow and low and the sounds commonly broad, shallow, and irregular in outline, and pass here 1 E. A. Smith, The Underground Water Resources of Alabama, Geol. Surv. of Alabama, 1907, pp. 250-251. ATLANTIC AND GULF COASTAL PLAIN 523 and there into grass-grown marshes, landlocked bays, and tidal flats. The sand reefs commonly lie ro, 15, or 20 miles off shore. The bays and marshes are partly bounded on the Gulf side by low sand banks, _ while between the bays friable sands and loams stand in vertical cliffs 5, 10, or 15 feet high. Instead of the high grounds and the low grounds of the Carolinas the entire surface of this portion of the alluvial valley of the Mississippi is low ground. The savannas are the most prominent features — broad tracts bounded by low scarps sloping steeply down and overlooking swamps, bays, and sounds. About the margins of the savannas are a certain amount of shrubbery and forests of pine or mag- nolia, but their interiors are often broad and imperfectly drained tracts of flat grass-land. The swamps are covered with reeds, sedges, and coarse swamp grass on their coastal sides, with live-oak groves on the coast rivers, and with canes and tangled shrubbery toward the interior.! The border of the higher portion of the plain on the inner margin of the coastal swamps is often a confused belt of knobs, crests, divides, spurs, peaks, and buttresses smoothly rounded, divided by flat-bot- tomed valleys with innumerable ramifications. On the floors of the valleys streams wander through broad plains of sandy alluvium. The summits of the hills and knobs reach elevations of 200 feet above tide and the valley flats to half that height. The upper plain represents the older surface, the lower the younger, and it is evident that after the excavation of the valleys and gorges depression ensued and the ravines were clogged with débris washed down from the hills, the final episode being an uplift of the land sufficient to drain but not deeply to erode the savannas or grass lands of the valley flats. The 600-mile western border of this district is a line of bluffs of com- - plex character which marks the border of a broad terrace, the bluffs consisting of a series of truncated spurs separated by ravines and broader valleys. At Memphis the bluffs are rounded and about too feet in height. At Yazoo and at Vicksburg they are 200 feet above the Mississippi flood plain. The slope of the Coastal Plain of Tennessee is eastward from the Mississippi bluffs, so that the highest portion of the Coastal Plain is immediately on the bluff, a physical feature which in the early settlement of the region determined the position of the roads, for it was the highest and driest portion of the country. The outer bor- der of the district, the edge of the Mississippi flood plain, undulates very gently in long low sweeps sculptured into a labyrinth of rounded hills 1 For a description of the shell hammocks and the coastal marshes, the character of their soil. etc., along the coast of Mississippi and Alabama, see E. W. Hilgard, Agriculture and Geology of Mississippi, 1860, p. 373 ff. 524 FOREST PHYSIOGRAPHY and long low valleys with a local relief not exceeding 200 feet and some- times as low as 50 feet. One of the most important physiographic conditions of this section of the Coastal Plain is related to the abandoning of the old fields and the removal of the forest cover. It appears that in a state of nature the drainage was accomplished without unduly rapid dissection of the fertile surface soil—a yellow loam from 3 to 7 feet thick. But when the oak forests were removed in the settlement of the country and the plantations abandoned during the Civil War, the hills, no longer pro- tected by a forest foliage, and the soil, no longer bound by the forest roots, were vigorously attacked by the streams and gullied and chan- neled in all directions. Year by year .the formerly fertile fields were invaded by gullies of ever-increasing width, the soil of long geologic growth disappeared down the stream-ways, and the land was gashed and harmed beyond belief. “The washing away of the surface soil . . . diminished the production of the higher lands, which were then commonly ‘ turned out’ and left without cultivation or care of any kind. The crusted surface shed the rain water into the old furrows, and the latter were quickly deep- ened and widened into gullies — ‘red washes.’ . . .” “As the evil progressed, large areas of uplands were denuded completely of their loam or culture stratum, leaving nothing but bare, arid sand, wholly useless for cultivation; while the valleys were little better, the native vegetation having been destroyed and only hardy weeds finding nourishment on the sandy surface. “Tn this manner whole sections, and in some portions of the state [Mississippi] whole town- ships of the best class of uplands have been transformed into sandy wastes, hardly reclaimable by any ordinary means, and wholly changing the industrial conditions of entire counties, whose county seats even in some instances had to be changed, the old town and site having, by the same destructive agencies, literally ‘gone down hill.’”’ + ; Specific names have been given to the erosional features of this district: a “break’’ is the head of a small retrogressive ravine; a ‘‘gulf”’ is a large break with precipitous walls of great depth and breadth, commonly being one hundred or one hundred and fifty feet deep; a “gut” is merely a road-cut deepened by storm wash and the effects of passing travel. MISSISSIPPI VALLEY SECTION The sixth division of the coastal lowland is the great flood plain, or delta and flood plain combined, of the lower Mississippi. It is bounded on the east by a continuous line of bluffs and on the west by a conspic- uous ridge known as Crowley’s Ridge (north) and by an alternating line of irregular bluffs and valleys (south). This alluvial district is one of the most extensive of the really low areas of the continent, lying prac- tically at base level. In the southern part of the district the surface approximates tide level and indeed the outer border consists of perma- nent tidal marshes. The surface is very ill-drained, and bayous, lakes, and abandoned channels constitute an irregular maze of water upon a 1 E. W. Hilgard, Soils, 1906, pp. 218-2109. ATLANTIC AND GULF COASTAL PLAIN 525 plain with scarcely perceptible slope. Between the waterways are ridges of slightly higher land, some of which are the natural levees of channels long since abandoned. In the western part of the district the inter- stream areas lie so high that they are invaded only by the highest floods, but the surface material is fine and compact and the surface itself ill- drained; consequently the trees are either drowned by the floods or with- ered by the sun in the droughts and the surface is without a forest cover. It supports a patchy growth of coarse grass, the patches being known as the ‘‘black prairies” of southern Arkansas and Louisiana. In the northern portion of the lower alluvial valley of the Mississippi extensive land tracts were converted into lakes, flowing rivers trans- formed into stagnant bayous with uplifted areas between, and some South West Pass Fig. 209. — Finger-like extensions of the Mississippi delta. The entire area shown as land is swamp-land. stream courses actually diverted during the series of earthquakes between 1811-1813. This region includes the “sunk country” of Missouri and Arkansas and the “Reelfoot Lake district” of Kentucky and Tennessee, and forms the uplifted land of Lake County, Tennessee, one of the few sections of the Mississippi flood plain that escape the inundations of the highest floods. Between Lake Borgne and Mobile Bay the Gulf is advancing upon the land so rapidly that the coastal rivers are nearly submerged and occur as narrow mud banks like the Chandeleur Islands, or as com- 526 FOREST PHYSIOGRAPHY pletely submerged bars and shoals that parallel the coast. A further effect of submergence is shown in the excessive breadth of the lagoons. On either side of the immediate delta of the Mississippi the alluvium deposited by the great river has been modified by waves and currents into bars and reefs and the shore has a smoothly trimmed effect in contrast to the finger-like extensions of the actively growing portion of the delta. In the latter case the rate of deposition exceeds the rate of current wear and the land is being built seaward.! Meander development and sidewise bodily movement of the river as a whole have opened up a river lowland so flat as to make the coastal plains appear as uplands when viewed from the river. This lowland, the lower alluvial valley of the Mississippi, is from 30 to 60 miles wide and about 600 miles long. Over its flat surface the great Mississippi meanders in an extremely indirect course. Receiving the contributions of a vast network of tributaries themselves major streams, it is little wonder that the river is subject to numerous floods transmitted from the tributaries. Formerly they inundated a vast expanse of lowlands and were of yearly occurrence; now a system of restraining dikes or levees and expensive revetments partially restrain the great river, and floods are in a measure under control. At present the levee system com- prises about 1500 miles of structure and is about 71% completed. The number of square miles of overflowed land in 1903 was half the mileage for 1897. In 1882 there were 284 crevasses recorded; in 1903 there were only 9 of importance.” The extent of alluvial bottom land that escapes inundation is ex- tremely limited. The higher tracts are confined chiefly to the river border, hence this is where the principal plantations are located. From the low natural levees the land slopes gradually away on either side to swampy back country covered with heavy forests of cypress and gum. The first problem in the utilization of these back swamps is drainage, and until the present canal system is perfected and extended but little development of the forest can be expected. The possibilities are suggested by the large shipments of logs, staves, headings, etc., from the Yazoo basin, where there was practically no lumber industry 1 The Mississippi is estimated to carry to the Gulf of Mexico enough land waste to cover a square mile 268 feet deep, an amount of material that would require a train of 44 loaded cars arriving at the Gulf every minute, the specific gravity of the sediment being taken at 2.5 and the capacity of a railroad car being 50,000 pounds. (J. E. Carman, The Mississippi Valley between Savanna and Davenport, Bull. Ill. State Geol. Surv. No. 13, 1900, p. 23.) 2 R. M. Brown, The Protection of the Alluvial Basin of the Mississippi, Pop. Sci. Mo., Sept., 1906, pp. 248-256. See also W. S. Tower, The Mississippi River Problem, Bull. Geog. Soc. Phil., vol. 6, 1908, pp. 83-100. ATLANTIC AND GULF COASTAL PLAIN QR: 90. y: = Z 9 Sai: = = fee) — — Memiphi (a eee eet se — amy ae 3 SS = Zz H ye xz : oe == j= [= = y = FV sburg: = = = oy a) Re ¥- Semin \ Lak ake Ponfehartrain/Y < 2 38 79o ae Sg J =) > New Orleans ° of Se SS Mn x 10° 92 90 Scale of Miles 0 20 40 80 120 160 Fig. 210.— The lower alluvial valley of the Mississippi. 527 528 FOREST PHYSIOGRAPHY until the present levee system was constructed, the floods reduced, and swamp drainage begun. At the northern end of the lower alluvial valley of the Mississippi are two main belts of lowland separated by a long ridge of varying width. The two lowlands are joined across the ridge by a broad belt of lowland and by several narrow stream valleys. The lowlands are called the Advance (west) and the Cairo (east) lowlands, Fig. 210. The Advance lowland is essentially a unit, while the Cairo lowland is divided into two subordinate belts by a long ridge known as the Sikeston Ridge. The Advance lowland is from 2 to 8 miles wide and is bordered in many places by relatively steep bluffs. The surface deposits of the lowland are sand and clay. A few long, low, flat-topped, sandy “‘islands”’ lie upon it at various places and furnish the only bits of land favorable for agriculture. The Cairo lowland on the east is nearly level and is bordered by prominent bluffs. The surface deposits consist of sand and silt, and on both this and the Advance lowland are numerous sloughs containing varying amounts of water. The small streams draining both lowlands have banks so low that they overflow during ordinary floods and the land is often submerged. The ridge between the two main lowland belts varies in height, form, and material, and is broken into a number of isolated sections. The main ridge extends southwestward, and to it the general name of Crowley’s Ridge has been given. At its broadest part Crowley’s Ridge is about 20 miles wide. Its upper surface is generally in the form of a rolling plain sloping gently westward. All the creeks flowing upon it have broad flood plains, with high land only in the form of narrow ridges on the main and secondary divides. On the main divide the valleys are deep and narrow and the surface maturely dissected. The eastern side of Crowley’s Ridge is a definite and steep bluff throughout; the western edge is not so definite, only a part of its course being marked by bluffs. The western lowland (Advance) was once occupied by the Mississippi River. The eastern lowland (Cairo), once occupied by the Ohio River, is now occupied by the Mississippi. The change of course in the Mississippi was brought about through stream capture by tributaries of the Ohio working westward into Crowley’s Ridge. A second capture of the Mississippi took place at a later time and brought about a further adjustment. The length of the valley abandoned by the first change of the Mississippi was more than 200 miles; the second change effected an abandonment of about 50 miles. The second change took place so recently that the river has not yet had time to clear its channel of rocks, much less to form a flood plain. The first change in the Mississippi brought its point of junction with the Ohio to a short distance south of New Madrid, Missouri; the second change brought it into its present relations with that river. 1 C. F. Marbut, The Evolution of the Northern Part of the Lowlands of Southeastern Missouri, Univ. Mo. Studies, vol. 1, 1902. ATLANTIC AND GULF COASTAL PLAIN 529 LOUISIANA-TEXAS SECTION The next section of the Coastal Plain extends from the Mississippi flood plain to the Rio Grande; the coastal margin of this section con- sists of wave-built ridges or reefs of exceptional continuity, Padre Island, south of Corpus Christi Pass, being 100 miles long. The lagoons back of these reefs are equally continuous; light-draft vessels may sail from the mouth of the Rio Grande through the Laguna de la Madre and through shorter sounds to Matagorda, 250 miles away. The depression of the coast is so rapid that the sand reefs are drowned and the lagoons are increasing in breadth. The combination of lagoons, fringing reefs, drowned river courses, complex est- uaries, and clean sea cliffs is an ex- pression of a sea advancing on a low-lying land, and the width of the belt in which these features are de- veloped is in a general way propor- tional to the rapidity of depression. The border of the plain is made up of alternating savannas and swamps or shoal bays. The low grounds are abandoned to reeds and sedges. The savannas are so low as to be clothed only with coarse grass and dotted with scrub pine or palmetto. Broad, low, natural levees like those of the Mississippi occur throughout the savannas, and these are commonly wooded, while the interstream tracts are prairie lands. Narrow belts of forest thus occur along the waterways except Fig. 211. — Coastal features of Texas, long, simple sand reefs enclosing narrow lagoons. where the surface has been cleared for agriculture. Along Red River are well-wooded tracts with oak and hickory on the uplands, poplar and liquid amber on the lowlands, cypress and tupelo in the swamps. In the coastal plain of Louisiana the rivers are steep and sluggish, such as Vermilion Bayou, Calcasieu River, etc. Each of these streams is narrow, deep, and clear, has scarcely any current, expands into a broad shallow lake, and enters the Gulf through a shallow bay. All of them have features characteristic of drowned streams. In south- 530 FOREST PHYSIOGRAPHY eastern Texas the streams of the Coastal Plain are similar to those of Louisiana; but west of the Nueces the coast drainage fails almost absolutely, the whole stretch of the coast to the Rio Grande containing only two small creeks. Though the greater part of the Louisiana-Texas section of the Coastal Plain is crossed by large rivers, a considerable part of the surface is poorly drained. Water stands in a multitude of small lakes or ponds throughout the year and there are large tracts covered with water during the wet season. The innermost belt of country belonging to the Coastal Plain rises rather rapidly from the adjacent seaward belt and has a more broken surface with numerous, small, rounded hills. The general elevation of this belt does not exceed 175 or 200 feet above sea level. The surface is In general timbered.! SOILS AND VEGETATION In Texas the Coastal Plain consists of a western sandy subdivision and an eastern clayey subdivision; the line of separation is the Guad- alupe River in Victoria County. Soil distinctions while of great im- portance to vegetation in the interior of Texas are of little importance on the low Coastal Plain, where the chief consideration in tree growth is the water supply. The outer margin of the eastern and clayey subdivi- sion is swampy and flat and but little above sea level. Patches of sandy land occur on the borders of the clay area. Forests of pine, oak, and magnolia fringe the northern border of the coastal plain on the higher grounds, various species of gum occur on the benches, and heavy forests of black and red cypress are found on the low river flood plains. The greater part of the Texas section of the Coastal Plain consists of groups of prairies separated by forest tracts. One of the most interesting forest trees of the low coastal margin is the loblolly pine, which grows on slightly elevated mounds of the Texas lowlands where it forms forest islands. The surrounding prairie is covered with water several months each year and is wet nearly the entire year, so that loblolly pine develops almost undisturbed by fires. The gradual filling in of the prairie by the loblolly groves causes the land to become drier, and strips of young growth bridge the space between groves and finally develop into large bodies of forest. The seeds of the loblolly are scattered partly by the wind in a southeasterly direction © if they ripen at the end of September or early October when northwest winds prevail, and partly by the southeastward drainage toward the Gulf during periods of high water. The drying up of the prairie and 1 Hayes and Kennedy, Oil Fields of the Texas-Louisiana Gulf Coastal Plain, Bull. U.S. Geol. Surv. No. 212, 1903, pp. 10 ff. ATLANTIC AND GULF COASTAL PLAIN 531 the reclamation of the land is a natural process common to the whole coast of Texas and Louisiana. The swamp vegetation adds by its decay to the surface material and gradually the surface is built up- ward to the point where shrubs and trees come in. The process is rapid enough to cause marked changes in the distribution of timber in 40 or 50 years, so that the marshes within the loblolly pine forest of eastern Texas, which a few years ago were impassable, are now acces- sible to man and to cattle.! Southward from San Diego the Coastal Plain is composed of a belt of brown sand probably 25 miles in width. It is a rolling country more or less covered with mesquite and chaparral. Still farther south is a gray sand belt having a width of 50 or 60 miles and, except for a few live oaks, practically without trees. It has been called the Great Texas Desert, and across the face of it stretch two belts of moving sand hills in an east-west direction. Each belt consists of a double row of dunes from a half mile to a mile apart, moving westward under the prevailing easterly winds. Some of them (northern Star County) are from go to too feet high, but the size of the dunes decreases east- ward and near the coast they appear only as white spots lying but slightly above the plain. Some of them are almost circular with a central depression, others are oval, and still others are in the form of great crescents, the typical symmetrical dune shape. The dunes are composed of extremely fine sand of snow-white color, and the lightest wind sets the fine grains in motion. In several instances tall live oaks have been buried so deeply that only the dead tops of the highest branches indicate the fate of the groves invaded by the dunes.’ From one to two hundred miles west of the coastal tract the forests dis- appear or occur chiefly in the form of scraggly growths of blackjack and Chickasaw plum. Still farther west the mesquite is found in low orchard- like groves on the interfluves, with hackberry and acacia along the streams. Eventually these forms give place to the sage and the cactus of the deserts, except where the fertilizing streams have led to reclama- tion or support a more prosperous native growth. PHYSIOGRAPHIC DEVELOPMENT The structure and physiographic history of the Coastal Plain of Louisiana and Arkansas bear such an intimate relation to the major topographic features that a word in regard to them is necessary at this 1 R. Zon, Loblolly Pine in Eastern Texas, Bull. Forest Service, U. S. Dept. Agri., No. 64, 1905, Pp. 9-10. 2 Idem, p. 16. 532 FOREST PHYSIOGRAPHY place. In the Arkansas-Louisiana district the relatively soft formations of Cretaceous and Tertiary age which form the Coastal Plain overlie a peneplain developed upon older and greatly deformed strata. In Miocene and early Pliocene times erosion was active at a level distinctly lower than that of the old peneplain and resulted ultimately in the formation (late Tertiary) of local base-leveled surfaces, essentially coincident with the Coastal Plain and continuous with an uneroded outer area lifted so slightly above sea level as to suffer no important modification during this ero- sion period. This Tertiary base-leveling while extensive was not complete, and many rem- nants of the older and higher surface still project above the common level: After the partial development of a Tertiary peneplain there came a depression of the entire region of sufficient amount to allow the forma- tion partly by river aggradation, partly by marine deposition, of a great blanket of silts, sands, and gravels (the Lafayette formation), which still occurs widely distributed throughout the region, its materials forming the surface soil to a large extent.1 The process of excavation of the material constituting the Coastal Plain has gone on with but one interruption since the deposition of LN PT ck\rkana Z AW WWWIH/S SK RRS AWAY The Lockesburg. Saratoga, Sulphur, — ee Flood-plain and terrace areas. Fig. 212. — Prominent topographic features of the Gulf Coastal Plain in northern Louisiana and southern Arkansas. (Veatch, U. S. Geol. Surv.) and Kisatchie wolds- the Lafayette formation. Through either climatic change or crustal deformation or both the streams of the region after cutting out valleys began to aggrade their valley floors, but after the partial filling of the valleys reéxcavation was begun. The result is that the streams are 1 A. C. Veatch, Geology and Underground Water Resources of Northern Louisiana and Southern Arkansas, Prof. Paper U. S. Geol. Surv. No. 46, 1906, p. 46. . ATLANTIC AND GULF COASTAL PLAIN generally intrenched below the level of the upper surface of the alluvium which now occurs in terrace form fringing the borders of the valleys and constituting an intermediate level of bench land of considerable extent between the general surface of the Coastal Plain and the flood plains of the streams. Besides the terrace and flood-plain areas are extensive areas of upland that are interesting chiefly because of the development of alternat- ing belts of higher and lower lands in the form of cuestas and inner lowlands. The best de- velopment of these features is in Arkansas and Louisiana where there are four more or less per- sistent cuestas that follow the general strike of the formations. The three northerly ones are the Lockesburg, Saratoga, and Sulphur cuestas; the southernmost is distant from the other three and is called the Kisatchie cuesta. It is formed on the outcrop of the hard sandstone forma- tions known as the Catahoula (Oligocene), which have been brought to the surface and exposed by a fault, so that the cuesta is not a typical one but should be regarded rather as a modified fault scarp very similar, however, in its topog- raphy to a cuesta formed by ordinary differen- tial erosion, Fig. 213. The Sulphur cuesta is formed on the harder members of the lower Eocene, and the Saratoga and Lockesburg cues- tas on the Cretaceous. SPECIAL FEATURES MOUNDS Two important local features merit description at this point, for they constitute notable depart- ures from the normal types of topography and drainage with which we have so far had chiefly to deal. These are (1) the two types of mounds which exist in the district and (2) the lakes at the lower ends of the tributaries of the Red sanqsayooy i i LOOT ONAVT \ = WSs Zz 3 > WS Td LOU PI. SoZ BURYIeXIy, Sr SS wan anydingy qiodaacryg Adana su1goy SET i za WY) eZ Feet 500: 500 1000 1600: Scale of Miles 533 (Veatch, U.S. Geol. Surv.) The formations range in age from Paleozoic and Lower Cretaceous on the right rise. Note the structural variations and the topographic complexities to which they give Fig. 213. — Cross section of the Gulf Coastal Plain in Louisiana and southern Arkansas. to Tertiary (Oligocene) on the left. 534 FOREST PHYSIOGRAPHY River. The first type of mound occurs in valleys where erosion has revealed the presence of domes with steep marginal dips attributed to igneous intrusions which did not reach sufficiently far toward the surface to be exposed by erosion. Only the top of the dome has beén removed, and the topographic expression is commonly not a mound but a depres- sion rimmed about by harder limestone, from beneath which the softer formations have been removed by erosion. ~The second type of mound is not only a structural uplift but also a topographic elevation and has an extremely wide distribution, being well developed on the prairies and pine flats along the coast of Louisi- ana and Texas, where are found the ‘“‘pimple prairies” popularly but erroneously associated with the oil deposits. They occur irregularly throughout the Coastal Plain and are best developed along the river terraces, though they are frequently found on the upland as well. No satisfactory theory has yet been formulated to account for them. On account of the wide distribution and unsettled character of the problem they represent, a partial bibliography of references is given at this point for the convenience of the reader. The different theories are discussed at some length by Veatch,’ from whose paper the following list of refer- ences is taken.” RED RIVER RAFTS Peculiar interest attaches to the great rafts of the Red River Valley on account of the unique combination of conditions which they repre- sent and the important changes they have effected in the hydrography of the Red River and its tributaries. The name raft is applied to the natural accumulations of timber along the river caused by the caving of the banks on the outside of the river bends. The main raft of the river began or at least was first known at Natchitoches, a town located below the raft and at the head of navigation. This was known as the Great Raft, and it grew steadily upstream with the constant addition of 1 A. C. Veatch, Geology and Underground Water Resources of Northern Louisiana and Southern Arkansas, Prof. Paper U. S. Geol. Surv. No. 46, 1906. 2 John C. Branner, Science, n. s., vol. 21, 1905, pp. 514-515; E. W. Hillgard, idem, pp. 551-552; W. J. Spillman, idem, p. 632; A. H. Purdue, idem, pp. 823-824; and C. V. Piper, idem, pp. 824-825. Branner and Purdue suggest that these mounds may represent immense con- cretionary formations. Spillman refers certain mounds in southwest Missouri to unequal weathering of limestone containing large chert masses. Branner gives many references to the mounds of the Pacific coast, for which he states the following theories have been advanced: (x) surface erosion, (2) glacial origin, (3) eolian origin, (4) human origin, (5) burrowing animals, including ants, and (6) fish nests exposed by elevation. D. I. Bushnell, Jr., Science, n. s., vol. 22, 1905, pp. 712-714, has suggested the human origin theory, and this phase of the matter has been discussed by Veatch, Science, n. s., vol. 23, 1906, pp. 34-36. Fig. 214. — One of the timber jams composing the great Red River raft. In such a jam silt accumulates very rapidly and effectually fills the channel. (Veatch, U.S. Geol. Surv.) Pc oe Jue ake, Cannisnia \ Old Lake iy 4 cusiz py nT 777 Scale of miles Orn 25 Fig. 215. — Lakes of the Red River valley in Louisiana at their fullest recorded development. (Veatch, U. S. Geol. Surv.) P. 535 536 FOREST PHYSIOGRAPHY Fig. 216.— Timber deadened in temporary raft lake which was drained by the removal of the raft. (Veatch, U. S. Geol. Surv.) Fig. 217. — One of the Red River rafts after partial recutting of filled channel, 1873. (Veatch, U. S. Geol, Surv.) ATLANTIC AND GULF COASTAL PLAIN 537 material in that direction until in the latter part of the fifteenth cen- tury it had reached a point near Alexandria. The effect of the natural dam which the raft created was to raise the level of the river on the upstream side of it and cause a ponding not only of the main river but SAlxids of Red River AS eA LEXANDRIA =) ee (e —— =| >a SEAS. COTES SECoulee=les Grues| 2s = oy ers = RAPTP—Cheneyvl le == 1905 92°20’ fe 92°00 « Fig. 218. — Showing diversion of Red River below Alexandria, La., and location of rapids. Map also shows typical drainage features in the Red River and Mississippi River flood plains. (Veatch, U. S. Geol. Surv.) also of the tributaries within the reach of the ponded portion of the main stream. In some cases the rise of water in the main stream was sufficient to cause it to discharge about the natural dam and through the timbered bottom lands on one side or the other. Driftwood would be quickly accumulated about the new point of discharge and again the 538 FOREST PHYSIOGRAPHY channel would be shifted. At the end of about 200 years (estimated) the lower part of the raft began to decay and the front of it to move upstream as a great irregular accumulation of log jams and open water about 160 miles in length. Its average rate of advance was about four-fifths of a mile a year during the period between 1820 and 1872, though the rate was intermittent and to a large extent dependent upon the discharge of the stream from year to year. As the head of the raft moved upstream it blocked all the tributary streams in succession and caused the formation of lakes at the points of junction with the main stream. In sim- ilar fashion the tributaries that were freed by the decay and retreat of the raft material along the front of the raft again discharged in the normal way and proceeded to dis- sect the deposits that had been accumulated on the floor of the temporary lake at their mouths. "Sarth A Since the removal of the rafts from the Red River the water has gradually resumed its old level. 5 10 From 1873 to 1892 the river had lowered its bed about 15 feet at a Fig. 219. — Growthand drainage of the raft lakes at the Arkansas- point above Shreveport; the lake Louisiana state line. (Veatch, U. S. Geol. Surv.) areas are rapidly draining and the extensive lake system along the course of the Red River shown on most maps is practically nonexistent. The continuance of some of the lakes and the occurrence of rapids and small falls in some of the tributaries are due to the drainage deflections that resulted from the silting up of the floors of the temporary lakes and the assumption of a new channel by the recreated stream. Superposition was inevitable under ATLANTIC AND GULF COASTAL PLAIN 539 these circumstances, the stream in certain instances flowing over projecting and now covered Tertiary spurs that have been revealed by erosion. The most notable instance of such diver- sion is in the case of the Red River itself, which has rapids immediately above Alexandria. The main features are shown on Fig. 218. SOILS Although the soils of the various districts of the Coastal Plain have been incidentally mentioned in the discussion of topographic and drain- age features, a connected description of their qualities is essential in understanding their geographic distribution and origin. The soils of the Atlantic and Gulf Coastai Plain are for the most part composed of sands and light sandy loams, with occasional depos- its of silts and heavy clays. The heavy clays are found principally near the inner margin of the Coastal Plain. The silts, silty clays, and black calcareous soils upon which the rice and sugar-cane industries of southern Louisiana and Texas are developed have no equivalents in the Atlantic division. Differences in the method of deposition, subsequent erosion, and drainage conditions are responsible for a great diversity of soil types with complicated relationships. The following are the most important series that have so far been recognized: (1) Light- colored sandy soils underlain by yellow or orange sand or sandy clay subsoils. Where the drain- age is insufficient, the subsoil is often mottled. With few exceptions these are special-purpose rather than general farming soils, and constitute the most important truck soils of the coastal plain. (2) Dark-gray to black surface soils, underlain by yellow, gray, or mottled yellow and gray subsoils. The dark color of the soils is due to an accumulation of organic matter during an earlier or existing swampy condition. This series is intermediate between the light-colored soils on the one hand and the peat and swamp areas on the other, and occupies depressed areas, or areas so flat that the water table is at or near the surface, except where the country is artificially drained. In this series the fine sandy-loam type supports a heavy growth of cypress, gum, magnolia, and other water-loving trees and undergrowth. It is characterized by level or slightly depressed surface features. Lack of drainage is responsible for the ex- istence and peculiar characteristics of the type. In most cases artificial drainage is imprac- ticable, owing to the low gradient. (3) A third series is derived largely, but not entirely, from the Lafayette mantle of gravels, sands, and sandy clays. The surface soils are usually gray to brown in color and are invariably underlain at a depth of 3 feet or less by a red or yellowish-red sandy clay. The prevailing red color of the subsoil is the characteristic feature. (4) A fourth series includes the barrier islands or bars, shore-line deposits, and low-lying marshes of the immediate coast line. The barrier bars consist of sand accumulated by wave action and further modified by winds. The soils of the marshes, consisting of sandy loams, loams and clays, have been built up by the deposition of silt and clay carried in by streams, by wind-blown sand from the ad- joining sand areas, and by the decay of coarse salt grasses and other native vegetation. The agricultural value of these lands is very low, depending mainly upon pasturage and the coarse hay, and they are a distinct menace to health, as they form the breeding places of disease- carrying insects. Efforts to drain and reclaim these marshes have been attended with some success. The possibilities of successful reclamation depend upon the keeping out of the tides and the subsequent efficient drainage of the land. (5) The soils of the black, calcareous prairie regions of Alabama, Mississippi, and Texas are characterized by a large percentage of lime, especially in the subsoil, which in some of the 540 FOREST PHYSIOGRAPHY types consists of white, chalky limestone. They have been derived from the weathering of calcareous clays, chalk beds, and ‘‘rotten”’’ limestones (Cretaceous). In some localities remnants of later sandy and gravelly deposits have been mingled with the calcareous material, giving rise to gravelly and loamy members. ‘The soils are very productive and are at present devoted chiefly to the growing of cotton and corn. (6) The next series includes dark-gray soils found upon gentle slopes or undulations adjacent to streams and upon level or depressed areas in the uplands. Their formation is due largely to the peculiar topographic conditions resulting from the sinking of the limestones which underlie, in some of the areas, the materials from which other soils have been derived. They may be considered as colluvial soils formed by the creeping or washing of material from higher- lying areas. The sandy type has a considerable admixture of organic matter and lies on gentle slopes or undulations adjacent to streams. It is mainly hammock land, supporting a growth of hardwood forest, and is very productive. (7) A series consisting of gray and brown surface soils underlain by heavy, plastic, red, mottled subsoils. Where the basal clays are exposed by erosfon they show brilliant colorings, often arranged in large patches of alternating liver-color, red, and white. These clays are remarkably plastic and constitute the oldest marine deposits along the inland margin of the Coastal Plain. The soils are usually of low crop-producing capacity and are covered chiefly with pitch pine, scrub oak, and other trees of little commercial importance. TREE GROWTH OF THE COASTAL PLAIN The special vegetal features of the various sections of the Coastal Plain have been described in the foregoing pages. It remains to note certain general features more or less common to the whole province. Two points are of chief interest in this connection: (1) the effects of water supply and (2) the effects of texture and chemical properties on the native vegetation. Speaking in general terms one may say that the Coastal Plain exhibits (1) a number of inner belts more or less clayey in character and heavy, (2) a broad expanse of sandy land forming the long outer slope of the plain and almost all of the so-called upland, (3) river bottom lands fringed in many places with terraces or higher lands called ‘‘ second bottoms” or hammocks, (4) a border of marshy land consisting of fresh-water marshes on the landward side and of salt-water marshes on the seaward side, and (5) a line of long, narrow reefs but little above high tide except where blown by the wind into higher sand dunes. The reefs are for the most part sandy, though coral reefs fringe a part of the coast of Florida and some of the reefs of the Gulf region are covered with shells accumulated by the Indians. The sand reefs gen- erally bear a growth of pine which may be of good quality as originally on the reefs of North Carolina, or it may be stunted and mixed with low, gnarled cedar, etc., as on many of the reefs of New Jersey and Texas. The shell hammocks of Alabama and Mississippi are not re- stricted to the reefs but are found also on inlets and bayous easily 1 Soil Survey Field Book, U.S. Bur. of Soils, 1906. ATLANTIC AND GULF. COASTAL PLAIN 541 accessible to the water. Pitch pine, live oak, red cedar, sweet gum, and prickly ash are the most common shell-hammock trees. On the marshes and wet pine barrens the deciduous cypress (Tawx- odium imbricarium) and the common swamp cypress are the ordinary growths along a large part of the coast line. In addition many of the smaller marshes have a growth of stunted pine, maple, and black gum. The bottom lands along the rivers, the “ first bottoms,” generally have a stiff heavy soil with an excess of water and are subject to overflow. Their timber is generally luxuriant and consists of chestnut, white oak, sweet gum, black gum, magnolia, bottom white pine, cypress, black walnut, tulip, hickory, and ash. Among these the black and the sweet gums are most numerous and characteristic, hence the name “ gum swamps” commonly applied to the bottom lands. The second bottoms or hammock lands are slightly higher than the first bottoms, are never overflowed, have a silty not a clayey soil, and have a poorer tree growth, among which the most common types are white oak, water oak, bottom white pine, magnolia, ironwood, post and black-jack oaks, willow oaks, etc., all more or less stunted and scattered in their growth. The upland trees are economically of greatest importance and are at the same time most interesting from the ecologic standpoint. The great sandy expanses of the outer slope of the Coastal Plain are covered mainly with longleaf pine. The seaward margin of the sandy belt is, however, level prairie in most cases, on which the trees are disposed in clumps or groves. In Texas the islands of higher lands are covered with loblolly pine. In Louisiana groves of honey locust, red haw, and live oak dot the outer prairie region. Inland from the great longleaf pine belt the calcareous layers of the Coastal Plain outcrop and are covered with a distinctive and for the most part a lime-loving vege- tation. Oak and shortleaf pine are the most important types; mixed with them are red cedar, red haw, crab-apple, and honey locust. It is noteworthy that the heaviest clay soils on the inner margin of the Coastal Plain of Texas, Louisiana, Mississippi, and Alabama are marked by large numbers of prairie tracts alternating with groves of stunted trees, mostly red cedar, Chickasaw plum, and scrubby post and black- jack oaks. The most unproductive soils occur on the so-called ridge lands toward the inner margin of the Coastal Plain where some of the ridges are developed on the outcrop of sandy formations. Scarlet and post oak, black-jack oak, and stunted pines are the principal trees. They are characteristic growths on all the more infertile phases of the coastal- plain soils. The calcareous soils, by contrast, are notably fertile and 542 FOREST PHYSICGRAPHY have a good growth of hickory, ash, sweet gum, and honey locust. So faithfully do the distributions of these types follow the outcrops of the respective formations that a vegetation map on the one hand, and a soil map or a geologic map on the other, show striking correspondences in the positions of the boundary lines.' 1 For a description of the tree types of the Coastal Plain in Texas see R. M. Harper, Contr. Dept. Bot., Colum. Univ., Nos. 192, 215, 216, 1902-1905; for the timber belts of the Coastal Plain in Texas see W. L. Bray, Forest Resources of Texas, Bull. U.S. Dept. Agri., Bur. For. No. 47,1004; for the trees of Mississippi see E. W. Hilgard, Geology and Agriculture of Miss. 1860, and Soils, 1906, Chaps. 24 and 25; for Alabama see Charles Mohr, Plant Life of Ala- bama, Ala. Geol. Surv., 1901. CHAPTER XXVI PENINSULA OF FLORIDA GENERAL GEOGRAPHY Tue peninsula of Florida is remarkable for its projection southeast- ward into the Atlantic Ocean for 350 miles, its smooth, well-developed eastern shore line, its great keys and swamps, and its slight relief. It ranges in altitude from sea level to 200 feet above the sea at various points on the broad flat-topped tract which forms the center of the peninsula and to about 300 feet in the northwestern counties of Florida. A depression of 50 feet would cover all of southern Florida except the tops of sand hills and ridges, while an elevation of 50 feet would extend the shore line westward 20 miles from Cape Romano and make dry land of Biscayne Bay and Bay of Florida. The northern and western parts of Florida consist of a narrow, deeply eroded limestone (Vicksburg) upland which descends southward rather abruptly to a low coastal region and northward by more gentle descents to the adjacent Coastal Plain. The rivers of the northern region are consequent upon the initial slopes of the land as it emerged from beneath the sea except where they have removed the thin mantle of surface sand and superimposed themselves upon the older strata beneath. The southern and central portions of Florida form a great lake and swamp district, the lakes occupying sinks or depressions in the underlying limestones or shallow and broad depressions in the surface of the deposits overlying the limestones. Extreme irregularity of out- line characterizes the shores of the lakes and extreme irregularity of direction the courses of the rivers. Many streams of the peninsula have their sources in springs that occupy underground courses in the limestone. These bring to the surface large quantities of mineral matter, andit has been estimated by Sellards! that the rate of solution is sufficient to remove about 4oo tons per square mile per year, an amount which would lower the surface of the limestone about 1 foot in 5000 to 6000 years. The underground passages are sometimes several] hundred feet in diameter and several miles in length. The level char- 1 A Preliminary Report on the Underground Water Supply of Central Florida, Bull. Florida State Geol. Surv., No. 1, 1908, p. 16. 543 Montic Starke 544 Fig. 220. — Principal lakes and coastal features of Florida. PENINSULA OF FLORIDA 545 acter of the surface and the porous soil which mantles the bedrock afford an excellent opportunity for the formation of caverns, since they result in a large absorption of the heavy rainfall. The sink holes are in process of formation to-day and instances are known where sinks have been formed by the collapse of cavern roofs in different parts of the lake region. The same process has resulted in the formation of many natural bridges, as those of northern Walton County, at St. Marks River, and Santa Fé River. GEOLOGIC STRUCTURE In respect of structure Florida is an elevated crust block modified by a number of minor folds.'. The present peninsula is regarded as rest- ing on a much more extensive foundation of Eocene limestone, forming a plateau which formerly extended from the southeastern margin of the continent to Cuba and the Bahamas, and possibly to Yucatan. The isolation of the peninsula may possibly be due to faulting, and in part to current scour. In a number of places are gentle folds whose axes are generally parallel with the trend of the peninsula; these succeed one another in series between Lake Okechobee and Gulf of Mexico.2 One such fold is near the Atlantic coast, another near the Gulf coast, and a third in the vicinity of Brooksville and Plant City. The eastern ridge includes the well-known Trail ridge and forms the eastern bound- ary of the central lake basin. The western ridge forms the western boundary of this basin and passes through Lakeland.’ PHYSIOGRAPHIC DEVELOPMENT Recent stratigraphic studies have determined the fact that the Flo- ridian region was outlined in somewhat its present form in pre-Oligocene, probably Eocene time, but that it then existed in the form of a shallow submarine platform swept by ocean currents and blanketed by both. organic and terrigenous deposits. During the time between this early period and the final emergence of the crust-block as a peninsula the region was subjected to many changes of level —a series of four emergences and a corresponding number of submergences. During the progress of these changes the surface of the platform was never carried far below and never far above sea level. The submergences were usually about roo feet and never more than 200 feet, while the latter figure expresses the 1 W. H. Dall, Neocene of North America, Bull. U. S. Geol. Surv. No. 84, 1892, pp. 85-87. 2 Idem, p. 88. 3 Matson and Clapp, 2d Ann. Rept. Fla. Geol. Surv., 1909, p. 48. 4 T. W. Vaughan, Sketch of the Geologic History of the Floridian Plateau, Science, n. s., vol. 32, 1910, pp. 24-27. 546 FOREST PHYSIOGRAPHY maximum elevation above sea level, a maximum that was attained during the Pliocene emergence. It was the fourth Pliocene emergence that gave the peninsula the outline that it has to-day. In the time since then the living coral reefs have developed, the Everglades have been formed, and the shores given their detailed characteristics. The net result of all the changes of level has been to leave the eastern side of the peninsula higher than the western and to bring into existence a number of small low folds whose axes run north-south in sympathy with the main axis of the peninsula. The Kissimmee River for example occupies an anticline flanked on both sides by low anticlines. TOPOGRAPHY AND DRAINAGE Although the mainland of Florida has slight relief it exhibits a con- siderable variety of topographic types, and since even small differences of elevation have a marked effect on the vegetation, it is possible roughly to divide the mainland of the peninsula into districts whose topographic features are intimately related to the vegetal growth. On this basis Florida may be divided into pineland and swamp. The pineland in- cludes ‘hammocks’ — isolated elevated patches supporting hard- wood trees of several genera — and many prairie or grassy tracts. The swamps include the marshy borders of the inland lakes and the coastal swamps with their sedges and black or red mangroves. It must be re- membered, however, that this line of demarcation between swamp and pineland is extremely irregular and, on account of the low relief, varies with the seasons to a certain extent. The total area covered by pine forests is probably in round numbers 300 square miles; on the east coast they extend in a narrow belt be- tween the Everglades and the coastal swamp from northern Palm Beach County to near Homestead. This belt is about 20 miles wide at the north, 6 miles wide near Jupiter Inlet, and from 2 to 8 miles wide towards the south. In northern Monroe County the pines grow in disconnected areas alternating with stretches of cypress... The pine- lands may be divided according to the relief into dunes, rolling sand plains, flat land, and rock ridges. DUNES The dunes of southern Florida lie near the coast and occur as a dis- continuous series of irregular mounds and ridges separated by inter- vals of flat or gently rolling country or by stretches of shallow water. 1 The best brief description of the topographic features of southern Florida and the one on which the following paragraphs are chiefly based is by S. Sanford, 2d Ann. Rept. Fla. Geol. Surv., pp. 177-231, where the results of many other investigators are assembled. PENINSULA OF FLORIDA 547 They seldom reach more than a few miles inland and but rarely face the ocean. An important group of dunes are those that occur near Jupiter Inlet at West Palm Beach, and on the east side of Lake Osborn. On the west coast of Florida there are few dunes south of the Caloosahatchee River. The dunes are not in movement to-day, and there is no lee- ward march which overwhelms trees and threatens dwellings as at Cape Henry, Virginia, and on Hatteras and Currituck Banks in North Carolina. When cleared of pine timber and palmetto scrub they may be utilized in the growing of tropical fruits. ROLLING SAND PLAINS The rolling sand plains include sandy stretches of the mainland with broad swells ahd low ridges, the swells being occupied by shallow lakes or lagoons and wet prairies or cypress swamps. On the east these sand plains form a belt with a maximum width of 6 miles and extend from the north side of Palm Beach County nearly to the Miami River and merge inland into the monotonous level of the flat lands and prairies bordering the Everglades. On the seaward side they are bounded by swamps; on the east coast the higher ground and the ridges are frequently covered with a straggling growth of spruce pine. In the hollows are many fresh-water lakes, some of which are several miles long. Most of them are less than ro feet deep and some are so shallow that they entirely disappear during seasons of deficient rainfall. The rolling sand plains are believed to be the result of wind action and to constitute but a broader development of the present dune topography near the coast. FLAT LANDS The name “flat lands” is used to designate the imperfectly drained pinelands lying between the narrow belt of rolling sand plains and the Everglades and their bordering prairies. The soil is a white sand which bears a thin growth of pine trees, with many prairie expanses a mile or more wide. In the rainy season these prairies are shallow lakes. There are also a number of sloughs and shallow ponds that in places support good growths of cypress, the pine and cypress often intermingling in very irregular fashion. On the west side of the lower end of the peninsula the surface between the Everglades and the Gulf is even more monotonously level than on the eastern coast, and the re- lations of swamp and dry land are more irregular. Much pine occurs in patches and strips separated by cypress swamps, a combination often designated as ‘‘pine islands and cypress straits.’’ In places prairies are scattered through or fringe the pinelands, and some of them make excellent cattle ranges. 548 FOREST PHYSIOGRAPHY ROCK RIDGES This term is applied to those outcrops of solid rock that rise above the general flat expanse of country, though they may not rise more than 2 feet above the general level and probably in no case does their ele- vation exceed 35 feet. Their extent is estimated at 200 square miles. On the east coast the rock ridges are of oGlitic limestone and separate the great saw-grass swamp of the Everglades from the fringe of man- grove swamps and salt prairie along the western shore of Biscayne Bay. Even in the Everglades some of the keys have a rocky foundation, but the only ones which expose bare rock are Long Key and those related toit. On the west coast of Florida hard rock outcrops are more scat- tered than on the east coast, but cover a wider area, running through the pinelands in strips varying in length up to several miles. SWAMPS THE EVERGLADES The swamp land of southern Florida includes the great saw-grass morass of the Everglades, the cypress swamps about it, and the salt marshes and mangrove swamps of the coast. The most important swamp is of course the Everglades with its wide expanse of sedge, its broad strips of shallow water, its scattered clumps of bushes, its many islands, and its underlying limestone floor. The Everglades reach from Lake Okechobee on the north to Whitewater Bay on the south, and are 50 miles wide in their widest part. Their area is estimated at 5000 square miles.!. The border of the Everglades is well known, but the vast interior of water- and sedge-covered muck has been visited by few geologists and crossed by. none. The Everglades tract lies in a huge shallow sink, or a series of more or less connected sinks.? Countless shallow ponds of clear water are found in which grow bulrushes, lilies, and other water plants, saw grass, flags, and cane. The Seminole name for the Everglades is “Grassy Water.” Scattered here and there in the sea of grass are islands of bushes and trées, called keys, which owe their origin to accu- mulations of vegetable matter; and the slight relief of the region is brought out strikingly by the fact that such slight accumulations of material should produce islands of drier land. The whole region ap- pears to be very young; it is almost without soil or definite surface drainage. As a result of the slight relief there is no sharp dividing line 1 L. S. Griswold, Notes on the Geology of Southern Florida, Bull. Mus. Comp. Zodl., vol. 28, 1896. : 2 A. Agassiz, The Elevated Reefs of Florida, Bull. Mus. Comp. Zo6l., vol. 28, 1896, p. 3o. PENINSULA OF FLORIDA 549 between the Everglades and the surrounding country, and a difference of two feet in water level means the difference between shallow lake and dry land for hundreds of square miles. No two of the maps of Florida agree either in the number, position, or outline of the lakes of the Ever- glades because of the variation in these features with the height of the water. Much of the eastern and northern shore of Lake Okechobee is bor- dered by cypress swamps, some of these containing the tallest cypresses to be found in Florida. On the east the Everglades are bordered by Fig. 221.—Swamp lands of the United States with degrees of swampiness shown in two shadings. (Gannett, U. S. Geol. Surv.) prairie and cypress swamps; in a few places patches of hardwood grow on slight elevations on the western border of the Everglades. Here are low irregular elevations, measured by inches, that diversify the distribu- tion of water and sedge; here, too, are narrow winding sloughs some of which extend for miles. Lake Okechobee itself is drained by a canal through the saw grass to Lake Hicpochee and the Caloosahatchee River. It is also drained by a few short streams that flow southward from the southern edge of the lake but are closed up after a few miles by thick growths of saw grass. The flatness of the Everglades may be appreciated by the elevations as determined at different times, which run from 6 to 23 feet in various portions, and it has been stated that the damming up of the main canal on the west of Lake Okechobee by raising the water three feet inundated the marginal prairies on many of the east coast ridges and seriously hindered the growing of vegetables. Bedrock lies at or near the surface toward the edge of the Ever- glades, and near Miami the rock forms bare ridges with a maximum height of 15 feet above the sea. The depth to bedrock in the central portion of the Everglades is 3 or 4 feet, and the flat- ness is so great that the word “‘basin” is inappropriate in describing it. 550 FOREST PHYSIOGRAPHY COASTAL SWAMPS The coastal swamps include the wet lands along the lagoons or the so-called rivers back of the barrier beaches of the east coast. Many of the swamps on the east side support a scrubby growth of mangrove, but on the west side, especially in the Shark River archipelago and the southern part of the maze of land and water known as the “Ten Thousand Islands,” the mangrove forms a notable forest, the trees reaching to a height of 60 feet or more, with green smooth trunks 2 feet or more in diameter at the base and without limbs for 30 feet above the ground. The mangrove forest rises from the Gulf like a green wall and is one of the most striking features of the shore line of southern Florida. FLORIDA IS NOT A CORAL REEF The work of Agassiz and others has shown that the older and popular view that the peninsula of Florida is an elevated reef is incorrect; the thickness of the coral reef formed since Pliocene times is probably about 15 feet, as determined by borings at Key West in 1895. Beneath the coral reef are Pliocene rocks, and beneath these in turn at a distance of 700 feet are rocks of Eocene age. The coral reefs now visible on the seacoast have but a moderate development inland. An elevated coral reef of notable width constitutes the outer edge of the peninsula of Florida. It has been traced from 20 to 30 miles inland at some points, although in other localities it has a very much more limited occurrence. This great development of reef rock should not, however, obscure the fact that the interior of the peninsula is of totally different origin, consisting of limestone, chiefly of marine origin, although with certain eolian facies. In the shore zone in-which the reefs occur there are also found great patches of eolian rocks filling intermediate sinks and alternating with patches of reef rock of variable extent.! DRAINAGE FEATURES DUE TO KARSTING The drainage of Florida is characterized by the presence of large numbers of sloughs, shallow ponds, and lakes. The interior is chiefly swamp, Fig. 220, with no well-defined river systems or stream valleys, and in spite of the low elevation of the larger part of the peninsula the streams that flow from the Everglades into the Atlantic have rapids in their upper courses wherever bedrock occurs. The drainage irregularities 1 The most comprehensive account of the geology and topography of the keys of southern Florida is by A. Agassiz, The Elevated Reef of Florida, Bull. Mus. Comp. Zool., vol. 28, 1896, pp. 29-62. PENINSULA OF FLORIDA 551 are due in part to the irregularities of the beds that form the land sur- face and in part to their general features and slight elevation above sea level. Additional factors controlling the stream systems have been the extensive karsting! of the limestone areas that underlie so large a portion of the state, with the production of many sink holes, underground chan- nels, caverns, etc. The higher portion of the state is honeycombed with underground passages, but these decrease in number on approach to the coast. No small part of the irregularities is probably due to the irregular manner in which the grasses grow, which in turn affects the distribution of land waste. The extreme irregularity which characterizes the drainage features of the peninsula of Florida must therefore be ascribed to three causes; first, the youthfulness of the entire surface and lack of time inewhich the drainage might become organized and the lakes drained; second, to the extremely slight relief of the surface and the absence of any dominating slope, as in the Coastal Plain of the Atlantic region or the Great Plains of the central part of the United States; third, to. the influence of the dissolving action of the ground water which proceeds in detail with almost a total disregard of the surface of the slopes of the land. The result is that un- derground water passages are opened from sink to sink and from lake to lake, oftentimes in opposition to the surface slopes, as in the celebrated case of the Danube and the Rhine in Europe, in which the upper part of the Danube is to a large extent absorbed by porous lime- stones, which in turn deliver the water to underground passages that lead to the Aach, a tributary of the Rhine system. The result is that the upper Danube belongs almost wholly to the Rhine system.2 To add to the complexity of such regions the covers of the under- ground channels are dissolved by percolating water to the point of collapse, passageways are blocked, and the waters rise to the upper level of the obstruction. In brief, it may be stated that the irregularities are in large measure due to the combination of surface and underground drainage systems which originate in different ways and which to a large extent obey dif- ferent laws. COASTAL ISLANDS The islands that fringe southern Florida are of several types. Some are long, narrow, barrier beaches crowned with coconut palms and bordered by mangrove swamps; some are true mangrove islands that resist wave action; some are low-lying sand banks supporting a scanty growth of beach grass and weeds; and some are of rock that reaches above sea level, these being covered with scrubby hardwoods, palms, and pines. Within the outer chain of islands that fringe the mainland or dot the Bay of Florida are other keys in all stages of development, from banks below sea level to banks bare at low tide and covered with mangroves that arrest the movement of sand, seaweed, and driftwood, and contribute to the reclamation of the sea floor. The existing keys and islands south of the peninsula of Florida are regarded as once having been continuous or practically so. It ap- pears that a process of gradual disconnection has taken place between 1 The name applied to the process that results in a karst topography, i.e., a topography marked by sink holes, underground channels, vertical shafts, etc., as developed on many lime- stone tracts and typically in the Karst Mountains of Austria. 2 Petermann’s Mitteilungen, 1907, and Natur Wissenschaftliche Wochenschrift, 1908, No. 7. 552 FOREST PHYSIOGRAPHY the Florida keys and the mainland proper. It is ascribed to the erosive and solvent action of the sea. The sounds may have originally been sinks similar to those of the Bahamas and the Bermudas, the action of the sea having broken through the barriers separating the sounds from the ocean. Once channels were formed the action of the sea would increase their width and depth, and the sea would thus gradually en- croach upon the floors of the sinks, forming more and more distinct sounds out of them, or even huge open bays like Key Biscayne Bay. The keys forming the small group immediately east of Key West, Fig. 220, have a northwest-southeast alignment. They are not coral reefs but are underlain by odlitic limestone which has an irregular sur- face more or less covered with marl and calcareous sand. The breaches between the islands are due to tidal currents caused by differences in time and height of the tides of the Gulf and the Strait of Florida. The great curve of the sand reefs on the eastern coast is caused by longshore current action, the currents moving from north to south as an eddy between the Gulf Stream and the mainland. There is a bedrock floor of limestone in the Biscayne pineland on Long Key and on adjoining keys in the Everglades. The coast of this part of Florida is either stationary or sinking, but if sinking, the rate of depression is extremely slow. At the present time there is in progress an extensive reclamation of the sea by the accumulation of large quan- tities of shells of marine organisms and by the abundant organic life associated with the mangrove swamps of the coastal border.1 There is the utmost difference in the vegetal covering of the islands, due to slight differences in elevation and to differences in the character of the surface. In some places it is bare rock, in others rock with a thin veneer of sand, marl, and leaf mold, and in still others a cover of cal- careous sand. On low land near the water’s edge mangroves are found, on the beaches coconut palms, on the low mar! flats grasses and sedges, while the higher ground, called “hammock,” supports a dense growth of scrubby hardwood trees, buttonwood, ironwood, madeira, etc., while on three keys patches of pine are found. The trees seldom reach a height greater than 20 feet. SOILS AND VEGETATION? The widespread occurrence of Pleistocene sand as a surface deposit a few feet thick results in an intimate relation between the soils of the 1 For an excellent description of these mangrove swamps, their great importance in the winning of lands from the sea, and their peculiar forms of vegetation, see N. S. Shaler, roth Ann. Rept. U. S. Geol. Surv., pt. 1, 1888-9, pp. 291-295. 2 The common and much more abundant types of vegetation are described in detail in connection with the topographic types with which they are so intimately related, pp. 546 to 550, PENINSULA OF FLORIDA 553 peninsula and this formation. Where erosion has been exceptionally active both the Pliocene and the Pleistocene deposits have been removed, leaving older geologic formations to form the soils, but such occurrences are relatively rare and unimportant.! In the southern part of Florida and in isolated patches elsewhere in the state peat and muck soils occur. These have their greatest development in the Everglades. They consist of organic matter mixed with a certain amount of sand and clay and are of recent origin. Their occurrence is confined to low up- land areas with imperfect drainage. In the upland portion of the state the Lafayette formation is found as isolated areas but these are of little importance except as tobacco soils in the northwest. At various places along the east coast Pleistocene marls and coquina in a more or less decomposed state form the subsoil, and the same is true on the west coast south of Bradentown. The clay and loam soils cover a very limited portion of Florida and are not of much importance. Clay soils are confined chiefly to small areas along the stream courses and are not tilled. The greater part of Florida has a sandy or sandy loam soil of low natural productivity but quickly responding to proper treatment. It is naturally deficient in moisture in spite of the rather abundant rainfall, and irrigation is practiced in a few places.2 Extensive drainage operations are in progress in the Everglades region which will ultimately mean the reclamation. of large tracts of peat soil whose natural pro- ductivity is high and which is suitable for the production of a large variety of tropical fruits. Florida extends so far southward as to exhibit subtropical or antillean forms of vegetation. The subtropical belt encircles the southern half of the peninsula from Cape Malabar on the east to Tampa Bay on the west.2 The royal palm, Jamaica dogwood, manchineel, mahogany, and mangrove are among the uncultivated tropical plants, and the banana, coconut, date palm, pineapple, grapefruit, and cherimoya among the cul- tivated plants of this district.* It is of interest to note here that the subtropical region enters the United States at two other points besides Florida — the lower Rio Grande region in Texas and the lower Colorado River Valley in western Arizona and southeastern California. 1 Matson and Clapp, 2d Ann. Rept. Fla. State Geol. Surv., 1900, pp. 37-43. 2 Idem, p. 45. 3’ C. H. Merriam, The Geographical Distribution of Life in North America with Special Reference to the Mammalia, Proc. Biol. Soc. Wash., vol. 7, 1892, p. 33. 4 C. H. Merriam, The Geological Distribution of Animals and Plants in North America, Yearbook Dept. Agri., 1894, p. 211; idem, Life Zones and Crop Zones of the United States. CHAPTER XXVII LAURENTIAN PLATEAU AND ITS OUTLIERS IN THE UNITED STATES LAURENTIAN PLATEAU TOPOGRAPHIC FEATURES. Tue Laurentian Plateau includes all the country north of Lake Superior, Lake Ontario, and the St. Lawrence line, as far as Hudson Bay and the Arctic shore of Canada; east and west it extends from Lakes Fig. 222. — Chiefly metamorphic and igneous rocks within the heavy lines, chiefly sedimentary rocks without; the Laurentian peneplain is dotted. (Wilson.) Athabaska, Manitoba, etc., to the Labrador coast... Its boundaries are indicated on Fig. 222 and include about one million square miles of land. It embraces practically the whole northeastern section of North America. Two outliers of the province lie in the United States: 554 LAURENTIAN PLATEAU AND ITS OUTLIERS 555 (1) the Superior Highlands of northern Wisconsin, northwestern Michigan, and northern Minnesota, and (2) the Adirondacks of northeastern New York, while it is practically continuous with the New England Plateau of related origin and character. Indeed the history of this great topo- graphic province is intimately associated with, and its forms are in many respects similar to, those of a considerable portion of the Appa- lachian region, and it is therefore necessary briefly to sketch its salient features. The Laurentian Plateau may be considered as a topographic unit, its chief features being those characteristic of a peneplaned region of crystalline rock — granites, gneisses, schists, etc. — bordered by over- lapping Paleozoic sedimentary layers which have been eroded into the form of an ancient belted coastal plain, then uplifted, extensively dis- sected by ordinary erosional processes in preglacial time, and, more re- cently, heavily glaciated. Two of the three great centers of glacial dispersion lay in the Laurentian province, the Keewatin and the Labra- dorian, and the effect of the great ice sheets that radiated therefrom was practically to denude the tract of its residual soils. With this set of physiographic conditions and changes we shall find associated practically all the topographic forms found in the region to-day.! The dominating feature of the Laurentian region is the remarkably even sky line whose character is maintained in spite of the great diversity of rock structures and hardnesses which occurs from place to place. Al- most anywhere.in the interior of the tract the horizon, as seen from an elevation, is almost as level as that of the sea and, also like that of the sea, is almost circular. The surface from point to point and in detail is decidedly broken, but in a general broad view it is strikingly flat and plateau-like. The degree of evenness may be appreciated from the fact that residual elevations of only 50 or too feet relative altitude stand out as prominent landmarks visible for many miles. The Laurentian area has been traversed by the officers of the Geological Survey of Canada in many different directions, and all the descriptions and photographs are alike in representing it as once having been one of the most typical peneplains in North America. Even in those parts of the now uplifted peneplain which have been most thoroughly dissected and glaciated, as in the region south of James Bay, the dominant topographic feature is still the even sky line intercepted only here and there by an occasional 1 A.W.G. Wilson, The Laurentian Peneplain, Jour. Geol., vol. 11, 1903. The true character of the great Laurentian area of eastern Canada was first set forth by G. N. Dawson, then Director of the Geological Survey of Canada, at the Toronto meeting of the geological section of the British Association in 1897, and in 1893 a systematic description of the region was prepared by Wilson. 556 FOREST PHYSIOGRAPHY isolated residual or monadnock which was able during the period of peneplanation to maintain a slightly greater elevation because of favor- able position on stream divides or superior hardness or both. The southeastern margin of the peneplain, where it borders the St. Lawrence River, is more rugged and uneven than the rest and is known as the Laurentide Mountains. Toward the extreme northeast and at the base of the Ungava peninsula there is a narrow range of mountains which extend to heights of 6000 to 8000 feet and stand prominently above the surface of the peneplain. The range includes the highest mountains Fig. 223.— The Laurentian Plateau as an uplifted, dissected, and glaciated peneplain in Labrador, east of Hudson Bay. Note the even sky line, the network of lakes, and the absence of a soil cover. (Low, Can. Geol. Surv.) in eastern North America. Broader and gentler undulations which give but slight diversity to the surface of the now uplifted peneplain also occur widely scattered in the form of low rounded domes and ridges roughly parallel to themselves and with their longer axes conforming in general to the strike of the rocks. Differences in elevation between different portions of the peneplain are roughly indicated by the numerous lakes, large and small, which occur at levels but little below that of the general surface. From the eleva- tions of these lakes it is found that the average gradient of the valleys is from 1 to 4 feet per mile. East of Cree Lake as far as the junction 1 A. W. G. Wilson, The Laurentian Peneplain, Jour. Geol., vol. II, 1903, p. 628. LAURENTIAN PLATEAU AND ITS OUTLIERS 557 of Churchill and Little Churchill rivers the surface has an average gradient of 1.8 feet per mile for 450 miles; on the Hamilton River the gradient for about 300 miles, partly above and partly below the Grand Falls, has a mean value of approximately 1 foot per mile. Considering the large number of falls and rapids in the courses of the streams drain- ing the Laurentian Plateau these values appear astonishingly small, especially when compared with such gently sloping surfaces as those developed in piedmont areas. The average gradient of the plains of Alberta is from 2 to 3 feet per mile, a value which distinctly exceeds that of the general slope of the surface in the Laurentian province, although we are accustomed to thinking of the latter as rugged and the former as smooth. The Laurentian peneplain has been dissected to such an extent as to be rough but not rugged, though even its roughness is not expressed in the figures representing the gradient of the hilltop plane. Although the derivation of the great Laurentian area from a pene- plain is clear, it is worth while to see that it now exhibits but few of the characteristics of a peneplain. A peneplain is an old erosion surface of slight relief which has been worn down by normal physiographic agen- cies almost to the level of the sea. The typical peneplain is one which also bears upon its surface an occasional residual elevation, is mantled by deep residual soils, and is absolutely without lakes, falls, rapids, or any other features indicating a youthful condition of the drainage courses. It is noteworthy that scarcely any of these qualities are pres- ent in the Laurentian area to-day. Instead of residual soils there are here almost no soils at all, only locally developed patches of glacial till or of water-laid material glacially derived. Instead of well-organized and low-gradient drainage courses, the Laurentian area exhibits the most irregular drainage features and is, par excellence, the lake region of North America; instead of standing near sea level the region is now at eleva- tions ranging from sea level to more than 2000 feet. Although it is one of the oldest portions of the continent, a large number of its fea- tures are due to a very recent geologic event — glaciation. In short, the Laurentian region is one which through unequal uplift and differ- ential stream and ice erosion has been so modified from its condition at the end of the preceding topographic cycle that its former peneplain character is recognizable only by the pronounced and uniform discord- ances that prevail throughout its entire area between general surface and geologic structure— a discordance that can be explained only by the assumption that it was once a region of prolonged erosion and that a pronounced, possibly a mountainous relief, was reduced practically to the level of the sea. 558 FOREST PHYSIOGRAPHY REPRESENTATIVE DISTRICTS LAURENTIDE MOUNTAINS With these general features in mind we may now turn to the more detailed features of representative portions of the Laurentian area, beginning with the Laurentide Mountains that border the St. Lawrence Valley on the northwest. Perhaps the most important point in this district is that relating to the deformation which the Laurentian pene- plain suffered in the uplifting process that raised it to its present level. The uplift was differential and is still in progress; the- general nature of the first uplift was a tilting of the whole northeastern corner of the continent upward at the north; the present movement is in har- Fig. 224. — The Laurentide Mountains north of St. Lawrence River at St. Anne de Beaupré. The lowland in the foreground is formed on limestone which terminates at the foot of the mountain slope. mony with and in continuation of the earlier movement. The Lauren- tide Mountains are due only in small part to their residual quality. They are due chiefly to the greater uplift of the peneplain along the border of the St. Lawrence Valley, so that its southeastern margin as viewed from the valley does not appear as the margin of a plateau but as a range of hills of sufficient height to be known as mountains. This marginal swell roughly parallel to the margin of the peneplain is succeeded toward the interior by a number of circular depressions oc- cupied by lakes, of which the St. John at the head of the Saguenay River is the most important. The streams of this portion of the plateau cross the marginal swell or the Laurentides through deep, steep- sided canyons of which the broadest and deepest is the fiord of the Saguenay. LAURENTIAN PLATEAU AND ITS OUTLIERS 559 HUDSON BAY-ST. LAWRENCE DIVIDE From the Laurentide Mountains northwest to Hudson Bay the Laurentian area is a plateau formed upon ancient crystalline rock with an average elevation of about 1500 feet above sea level. It rises slowly from 1ooo feet near the margin to about 2000 feet in the inte- rior. The surface is dotted with low rounded hills arranged in a series of ridges parallel to themselves and to the general strike of the rocks. These hills are the stubs of extensive and elevated mountain chains subjected to prolonged erosion which brought them down to their present state.! To the south of James Bay in the Nipissing and the Temiskaming regions the surface of the, plateau consists of a succession of more or less parallel rock ridges with intervening valleys occupied by swamps and lakes and with a general elevation rarely falling outside the goo to 1200 foot limits. There are no very pronounced hills, the highest seldom ‘attaining more than 300 feet above the adjacent country, while hills roo to 150 feet high are rather conspicuous topographic features. The areas of Laurentian granite and gneiss in the southern and southeastern portions of the district have weathered into a monotonous succession of hills and intervening rocky valleys; in the northwestern part of the Nipissing area there are more important elevations owing to the presence of resistant quartzite in country rock consisting of granite, diabase, and slate, the last-named member of the series weathering into a characteristically flat surface.’ LABRADOR PENINSULA East of Hudson Bay is the Labrador peninsula, a rough tableland having an elevation of about 7oo feet above sea level near its margin and about 2000 feet in its higher interior portions.’ The interior of the Labrador peninsula is so nearly flat that in an area of 200,000 square miles there is a difference in level of not more than 200 to 4oo feet, and the highest general level of the interior is less than 2500 feet above the sea. The divides are often ill defined and never prominent. 1 A. P. Low, The Mistassinni Region, The Ottawa Naturalist, vol. 4, 1890, pp. 11-28. 2 A. E. Barlow, Report on the Geology and Natural Resources of the Area included between the Nipissing and Temiskaming. Map Sheet, comprising portions of the district of Nipissing and of the county of Pontiac, Can. Geol. Surv., vol. 10, n. s., Rept. I, 1890, p. 21. 3 A. P. Low, Report on Explorations in James Bay and the country east of Hudson Bay drained by Big, Great Whale, and Clear Water rivers, Can. Geol. Surv., vol. 3, n. s., pt. 2, Rept. J, 1888, p. 16. 560 FOREST PHYSIOGRAPHY BARREN LANDS Likewise the “Barren Lands” in the far north of Keewatin, p. 565, although developed upon rock of most complex structure, are extraor- dinarily uniform in general elevation and in a broad view closely resemble the great plains of undulating grass-covered country of western Manitoba, Alberta, and Saskatchewan. Only here and there does a hard granite knoll projecting above the general surface indicate the character of the underlying rock and suggest the past history of the region. BORDER TOPOGRAPHY AND DRAINAGE Along the outer or southern and western border of the Laurentian peneplain is a group of land forms which represent on a great scale the general features of an ancient belted coastal plain. These features are best preserved in the region of the Great Lakes. The most continuous development of the cuesta feature is that associated with the outcrop of Niagara limestone which appears in the form of a pronounced but ragged escarpment extending from near Rochester on the southern shore of Lake Ontario diagonally northwestward through Ontario, form- ing the peninsula between Georgian Bay and Lake Huron. Thence it extends through Manitoulin Island and along the southern shore of the northern peninsula of Michigan. Farther west and southwest it con- stitutes the two peninsulas between Green Bay and Lake Michigan, and its final expression in the United States is in the southwestern corner of the Driftless Area, where it forms the boundary of an upland approxi- mately too feet above the general level of the country adjacent to it on the north.! West of Lake Winnipeg the Niagara escarpment again appears and fades out gradually to the north. Lakes Erie, Huron, Michigan, Mani- toba, and Winnipegosis are on the outer lowland of the ancient coastal plain; Lake Ontario, Georgian Bay, Green Bay, and Lake Winnipeg lie upon the inner lowland between the cuesta? and the old-land;? Lake Supe- rior occupies a depression not associated with either lowland or cuesta. Upon the Hudson Bay side of the Laurentian Plateau a similar margin of Paleozoic sediments may be seen, although they are more extensively concealed by the deposits of the present coastal plain than is the case 1 U.S. Grant, Lancaster Mineral Point Folio U.S. Geol. Surv. No. 145, 1909. See geo- logic map and descriptive text. 2 The term cuesta is applied to the culminating ridge on the inner side of a dissected coastal plain. Its outer slope is long and gentle, its inner slope short and relatively steep. The term old-land is applied to any extensive tract of older and higher land that supplied land waste out of which the young land, or bordering coastal plain, was formed. LAURENTIAN PLATEAU AND ITS OUTLIERS 561 on the southern margin. Over the peninsula southwest of Cape Henri- etta Maria in the angle between Hudson Bay and James Bay these Paleozoic sediments are developed to their greatest extent, but the cuesta feature is topographically very indistinct. On the eastern shore of Hudson Bay and James Bay the Paleozoic sediments have been dissected and submerged to the point where they now appear as chains of islands in some places continuous with anvil-shaped peninsulas. At one time the land stood higher than it now does, and dissection pro- gressed so far that with a later submergence of the region and the inva- sion of the eroded lowlands by the waters of Hudson Bay the cuesta was almost submerged; the degree of submergence was so great that the cuestas still appear as islands in spite of the great uplifts which the region has suffered in postglacial time. In a large number of instances the streams of the Laurentian Plateau discharge across its borders in gorges and canyons of note- worthy size and beauty. In places the gorges and valleys are steep- sided, narrow, and long; in other places they are short. They have their best development along the southeastern border, where the numerous streams that enter the Gulf of St. Lawrence from the northwest have deep canyons (sometimes with unscalable walls) cut to a depth of over tooo feet below the general level of the plateau. A specific case is the Saguenay, through which Lake St. John discharges into the St. Lawrence. Lake Temiskaming and Ottawa River, as far down as Mattawa, likewise occupy long narrow depressions cut beneath the surface of the plain. All of these canyons and deep, steep-sided valleys are marked by topographic unconformities, that is, well-defined shoul- ders where the steep slope of the valley wall intersects the rather flat slope of the upland surface, a clear indication that the canyon is of later origin than the upland surface below which it is incised. Where the deep preglacial canyons and gorges trend in the direction in which the ice flowed in the last glacial invasion the sides of the gorges have been steepened and the bottoms deepened. This is especially true of the Saguenay, which may probably be regarded as a typical glacial fiord, although it must be remembered that preglacial erosion had to a large extent prepared the valley for the maximum effects of glaciation as expressed in hanging valleys and deep channels below sea level. The fact that some of the long, narrow, and steep-sided gorges or de- pressions are at various angles with the direction of ice movement sug- gests that they owe their origin to the downfaulting of long narrow crust-blocks. This view would seem to be supported by the fact that Paleozoic sediments are preserved in the bottoms of many of the valleys. 562 FOREST PHYSIOGRAPHY The sediments in the Lake Temiskaming and Lake Nipissing basins are in valleys lying below the level of the plateau. The margin is well defined, often cliffed, and the tributary streams spill over the edge of the basin in a series of waterfalls and cascades where they pass from the more resistant to the less resistant rock. The long narrow depressions may be explained also on the assumption of an early Paleozoic period of valley cutting followed by a period of valley filling. In this view later valley excavation would proceed along the outcrop of the relatively softer filling in the older valleys. It seems quite probable that further work will show the validity of both explana- tions each being applicable to a number of valleys to the exclusion of the other. SPECIAL FEATURES OF THE NIPIGON DISTRICT The topography of the Nipigon area north of Lake Superior is of special interest because the normal erosion features have been either partly masked or greatly complicated by lava flows. To the northeast of Lake Nipigon is the Archzan old-land of crystalline rock containing numerous outliers of sedimentary strata. The southwestern portion of the present basin of Lake Nipigon represents a maturely dissected por- tion of the inner lowland of the ancient coastal plain. At a time when the relief of the coastal plain was at a maximum there occurred extensive intrusions of diabase and extrusive flows of basalt which to a large degree filled the inner lowland and even in some places poured out over the adjacent uplands. This period of lava flows followed the erosion that in turn followed peneplanation. Since its formation the diabase has been very extensively eroded so that only from 5% to 10% of it remains, but its occurrence dominates the region to such an extent that practically every prominent feature of the topography for one hundred miles along the southwest shore of Lake Nipigon and the adjacent country is associated with the trap sheets. Their upper surface is usually a tableland or mesa, and gorges have been cut to a large extent through the sheets by the draining streams. Beneath the basalt may be seen from place to place portions of the old peneplain which were still in existence when the flows occurred and which were covered over and so preserved by the resistant rock. The scenery is varied, bold, and picturesque, as determined by the sharp outlines of the igneous rocks whose cliffs were still further cleaned and freshened by glacial action. LAURENTIAN PLATEAU AND ITS OUTLIERS 563 CHANGES OF LEVEL The change of level which the Laurentian Plateau is now undergoing and which it has undergone in postglacial time is shown in a variety of most convincing ways. At Cape Nachvak, one hundred and forty miles south of Hudson Straits, are old strand lines reaching up to 1500 feet above sea level. Also on the east coast of Hudson Bay are well- preserved and numerous terraces cut in till and other deposits reaching to an elevation of 300 feet above the sea. Deep-water species of shells are washed out of the present beach and are found in stratified clays up to elevations of 500 ieet; and long, continuous lines of driftwood, chiefly spruce and cedar, occur 30 feet above tide. These data combine to prove that the region has suffered important changes of level so recently as still to preserve in the clearest manner the features indicat- ing uplift. Besides the physical features there are a number of impor- tant and convincing pieces of human evidence. The Eskimos of the region catch fish by constructing weirs of stone which at low tide impound fish that have come up close inshore during high tide for feed- ing purposes. These fish weirs and traps are found at all elevations up to 70 feet above present sea level, and as their use is invariably asso- ciated with the shore line of the time when they are employed, it is clear that the level of the land has changed not less than 70 feet since the Eskimos have inhabited the region. Hudson mentions win- tering in a bay full of islands where now only a canoe may pass, and the salt marshes about the border of Hudson Bay are drying up to such an extent that the geese and ducks which formerly made their home among them have extensively abandoned them, and the residents of the district find it increasingly difficult to secure the eggs of these wild fowl. Finally, the Indians often remark the growing distances between old buildings, forts, trees, and the shore line, although this piece of evi- dence is of course far less convincing than the others.! On the east coast of Labrador along the 1100 miles from St. John to Cape Chidley no careful measurements of postglacial uplift have been made, although Daly has approximated the amount by observations upon the distribution of bowlder clay and bowlders.? He concludes that the higher bowlder-covered zone has never been submerged since the ice sheet retreated from the country, and that the lower bowlderless zone is a wave-swept zone. He finds that the smooth unbroken surfaces 1 R. M. Bell, Proofs of the Rising of the Land around Hudson Bay, Am. Jour. Sci., vol. 1, 4th ser., 1896, pp. 219-228. 2 R. A. Daly, The Geology of the Northeast Coast of Labrador, Bull. Mus. Comp. Zool., vol. 38, 1902, p. 254. 564 FOREST PHYSIOGRAPHY of the roche moutonnées in the bowlder-covered zone (about 75 feet above sea level) contrast strongly with the jagged and riven wave-swept ledges below that level. Low, steep, and rugged cliffs are sometimes found associated with pebble and bowlder beaches of limited develop- ment. The elevatory movements continue in both Labrador and New- foundland, as determined by the testimony of the inhabitants in regard to the decreasing depth of water on the beach and the gradual shoaling of water over ledges and bowlders off shore. The recency of the shore uplift is also shown in the numerous fresh-water ponds lying back of the barrier beaches— ponds that at one time were true coastal lagoons on the landward side of submarine bars. The forms are strikingly fresh, the glaciated, bordering slopes having contributed but little land waste to shore filling. THE LAKE REGION The Laurentian area of Canada constitutes the greater part of the lake region of North America whose boundaries are shown on p. 565. It has been stated that there are areas within it 25% of whose surface is occupied by lakes.!. In order to get a closer value for this interesting condition the actual areas of the lakes represented upon a detailed map by Collins,” Fig. 225, were carefully measured. Some thirty independent measurements of the lake areas afford a mean that is close to accuracy. The result: about 16% of the total area of the district is covered with water. When it is considered that this value is for a representative dis- trict the figure is very striking indeed. However, even on this map only the more prominent lakes were represented on the original sheets. Were even the smallest lakes included, at least 20%, and possibly 25%, of the surface would be found covered with water. To secure a com- parable figure for total lengths of drainage systems composed of lake, about 2000 miles of river course were measured, with the result that the rivers examined showed lake in 57% of their total length. To a large extent the abundance of lakes in the prevailingly rocky Laurentian Plateau is due not alone to glaciation but to glaciation of a peneplain of such perfect development that the slightest degree of rough- ening by differential erosion would produce enclosed basins in large numbers. If the region were mountainous the effect of glacial erosion would be in the main to deepen the valleys and basins already formed and accentuate the preglacial differences of level and topographic form. We have only to recall the appearance of the glaciated mountain region of 1 A. W. G. Wilson, The Laurentian Peneplain, Jour. of Geol., vol. 11, 1903, pp. 645-650. 2 W. H. Collins, Report on a Portion of Northwestern Ontario traversed by the National Transcontinental Railway between Lake Nipigon and Sturgeon Lake, Ann. Rept. Can. Geol. Surv., 1908. 505 uRS 4 LAURENTIAN PLATEAU AND ITS OUTLII ‘ayes sty} jo dew v uo pozuasaidal oq JOU ULD yey} AlepUNo [eUOT}eUIOIUI 9Y} JO YNOS soy] [[eUs CoO‘Oz 4SRE] ye IV doy], “SOYL] JNOYIA sv olay UMOYS vaIe UL Jo J[vOS ISIy] v UO s|IeJop aSvuIvIp IOJ OFZ “Sty 99S “payeorpur ov saye] edrioursd 94} AjUQ — “Sze “Sry OT IN3AOVfaV AYLNNODS 3HL GNYV NVALVId NVILNSYNVT SHL : VOIMSWY HLYON J4O =—Ss BOR M7 eae = 00 Towwadns, i ES spooA{ a ld ( £0 OAV TA \ ~ ae / a5 Ne ir m™\~ 4. st a YQO07} ile Al NOISSY 3yXVv1 SHL hivg pnapbhwy q 209 566 FOREST PHYSIOGRAPHY the northern Cascades and the northern Rockies, with their strong topo- graphic discordances, to appreciate how different the Laurentian area would appear if its local relief had not been so slight as to be almost negligible. Even rocks of the same geologic age and general character vary sufficiently from place to place to cause important differences in _—~ aA \a = or S 6 e = a | \ ~ ones ¢ = ya ? pm) Sturgeon Lake ; 3 Seseganaga 150° / Lake _50°| I NETWORK OF LAKES f 1500 square miles within the Laurentian Pineplain, | | west of Lake Nipigon. Based on the National Sturgeon Transcontinental Railway surveys, 1906- Lake | 907307 Fig. 226. — Details of drainage in a limited portion of the Laurentian Plateau. (Based on Can. Geol. Surv. Maps and Railway Surveys by Collins.) topography; and when it is considered that so large a variety of rocks as granites, gneisses, schists, slates, and trap sheets that occur here extend these differences in resistance from place to place, it is clear that some roughening of the old peneplain must follow upon erosion of any sort and that in the case of ice the roughening would result in the formation of a very large number of lake basins. The preglacial slopes were almost never so strong as to overcome the general irregular scooping action of the ice, and every local overdeepening either as the result of decreased rock resistance or of glacial convergence produced reversed slopes that were occupied by lakes as soon as the glacial cover withdrew. LAURENTIAN PLATEAU AND ITS OUTLIERS 567 It is clear that the glaciation of the Laurentian Plateau would result in the formation of a large number of lakes, and that the lakes would be extremely irregular in contrast to the regularity of the lakes that occupy overdeepened portions of glaciated mountain valleys. Labyrinthine shores and shapeless islands are the rule in the Laurentian of Canada. The maze of waterways displayed by this region is the marvel of every traveler in it. The readiness with which some of the lake basins were formed may be judged from the fact that but aslight change of level may effect a change of outlet, as has been suggested many times in connection with the two-outlet phenomena commonly displayed by many of the Laurentian lakes. The feature was regarded as a novel one when first described, but the repeated finding of lakes with two outlets shows it to be a very much more general occurrence than was at first supposed. It has already been noted that the whole northeastern corner of the continent is suffering a change of level and that the movement is upward on the north. The result, in the case of lakes lying in depressions with low rims, is to cause a new outlet to come into existence as the old outlet ceases to function through uplift. The abandonment of an old outlet would of course be gradual and fora time two outlets would function. Even without such a change of level it has been thought that two-outlet lakes may have been caused by the formation of excep- tionally shallow basins on a peneplain surface. The two-outlet feature is indeed thought to be on such a scale as to be independent evidence of the tilting of the land. In some cases, how- ever, where two and even three outlets are in existence, they are found on the same side of a lake. Lake Chibougamou (50 miles south of Lake Mistassini), for example, is drained by two outlets both of which are on the east side.1 In such cases the two-outlet feature is unrelated to regional tilting and appears to be due instead to coincidence in the level of low cols on the rim of the lake basin. The condition is perpetuated by the hard rock which is but little affected by the clear streams. Besides the island-dotted lakes with ragged shores there is a distinct though comparatively small class of lakes that has more regular features. These lakes are usually longitudinal and are caused in a number of different ways. In some cases they owe their existence to the erosion of soft dikes, as, for example, the peninsulated northern shore of Lake Supe- rior with its long narrow bays and points. The dikes are commonly of greenstone, though sometimes they are of pegmatite, and in both cases are usually soft and their outcrop covered with lakes or streams. Some- times the correspondence of position of former dike and later river or lake is so close that where the dike narrowed the lake or river was narrowed and where it widened the drainage features widen. It must not be forgotten, however, that the dike material is not always softer than the country rock; where it is harder it stands out as a promi- nent ridge instead of a lake basin or a river valley. The straightness of trend of the dikes within short distances is directly related to the straight-line topography and drainage so characteristic in many portions of the area as to give a pronounced appearance of artificiality to it. Lake Temiskaming and the “Deep River” of the Ottawa drainage sys- 1 Data from Bateman, Can. Geol. Sury. 568 FOREST PHYSIOGRAPHY _ tem are examples of longitudinal lakes, but their origin may have been due to crustal warping or to block-faulting of the graben type. In still other cases the longitudinal lakes are due to blocking of valleys by glacial material. ; In addition to the lake types of the Laurentian that we have noted there is another class dependent upon broader and less local conditions. These lakes lie on the border of the region and their occurrence is coin- cident with that of large isolated areas of sedimentary rock within or on the border of the crystalline area. Whether these sedimentary masses represent downfaulted blocks of rock formerly more extensive Fig. 227. — View on the shore of Lake St. John, Quebec. The level sky line in the background represents the Laurentian Plateau; in the foreground is a lowland developed on limestone; in the middle distance are alluvial terraces. - than now, and preserved by downfaulting because carried below the gen- eral level of the peneplain surface, is not certainly known, but the explanation has a high degree of plausibility, for faulted zones have been found about the borders of some of them and everywhere the sediments are sharply localized and extend to great depths below the level of the ancient peneplain. Whatever their origin they were areas of weakness during the glacial period and indeed ever since the elevation of the peneplain, with the result that not only do lake basins lie in them but the outcrop of the softer sedimentaries is always marked by the existence of a lowland whose border is sharply de- termined by the crystalline rocks that rim about it. A typical occur- rence is at Lake St. John, which occupies a portion of an outlier of the Paleozoic sediments that border the St. Lawrence Valley on the north. The outlet of Lake St. John is over a harder rock rim to the LAURENTIAN PLATEAU AND ITS OUTLIERS 569 Saguenay, and falls and rapids mark the transition from one rock belt to the other. Hamilton Inlet, Great Bear and Great Slave lakes, and many other depressions have a similar origin, and in all cases small por- tions of the older sedimentary rock still cling to the border of the de- pressions in the crystalline tract, or, as at Lake St. John, occur in larger masses forming lowlands. Lakes are known to be ephemeral features of the earth’s surface and to be the easy prey of geologic processes. Commonly the streams that feed lakes bear such quantities of material into them and the drain- ing streams cut down the rim at the outlet so fast and vegetation ac- cumulates so rapidly that the life of a lake is in a geologic sense short. It is therefore interesting to find in this great lake region developed on the Laurentian of Canada a set of conditions that insure the longevity of the lakes within it, so that at least a whole geologic period would seem to be a conservative estimate of the length of time these lakes will persist. The principal conditions that are the basis for this conclusion are as follows. The fresh-rock rims have an exceptionally high degree of resistance. Instead of rotten rock the rock enclosing these lakes is firm. The products of decay were swept off by the continental ice sheet, leaving the undecomposed rock of the subsurface. Although falls and rapids abound the average gradient of the draining streams is only one to two feet per mile. In addition to the practical absence of land waste over vast portions of the Laurentian area, lakes are so abundant that the water is thoroughly filtered and the sediments are thoroughly entrapped. The streams are thus deprived of their cutting tools and waterfalls and rapids retreat with amazing slowness in the extremely hard rock. More powerful than these factors in extinguishing some of the Laurentian lakes is the vegetation that grows in such abundance about the borders of the shallower lakes as sometimes quickly to effect their reclamation. Lakes are turned into swamp or muskeg and then finally reclaimed, but the process is largely limited to the smaller and shallower lakes and does not affect to an important degree the conclusions we have just stated concerning the life of the lakes as a group. VEGETATION There is a limited forest growth in the hollows and along the borders of the Labrador region, but large portions of the interior are barren tracts of rock and muskeg. Exceptionally good forest tracts are found in a number of places. On the 51st parallel and east of Lake Mistassini, whence the Porcupine Range extends toward the north, is a well-wooded tract. The black spruce is here the predominating species. Other trees 570° FOREST PHYSIOGRAPHY are the white birch, poplar, willow, alder, and Banksian pine. The growth is surprisingly large; the diameters of the largest trees exceed 2 feet and the height attains 70 or 80 feet! Quebec and Ontario south of James Bay are densely forested, the forests extending westward to within a short distance of Great Slave Lake. Beyond this point the region may be described as a treeless moss-covered tundra whose subsoil is perpetually frozen. The northern boundary of the great transcontinental spruce forest closely follows the western shore of Hudson Bay from the mouth of Churchill River for a few miles, then curves gently inland; thence it extends northwesterly, crossing Island Lake, Ennadai Lake on Kazan River, and Boyd Lake on the Dubawnt. The next dividing point is just north of 60° on Artillery Lake. From this point the line curves south- westerly, crossing Lake Mackay south of latitude 64°. The banks of the Coppermine are the boundary to 67°. Tongues of timber follow the northward-flowing streams, with their warmer water, well into the Barren Grounds. The most remarkable stream of this kind is the Ark-i-linik, a tributary of Hudson Bay. From a point near latitude 623° north, within the main area of the Barren Grounds, a more or less con- tinuous belt of spruce borders the river to latitude 643°, a distance of over 200 miles by river. A few species of woodland-breeding birds follow these extensions of the forest to their limits. Alders occur in more or less dwarfed condition in favorable places well within the treeless areas, and several species of willows, some of which here attain a height of 5 or 6 feet, border some of the streams as far north as Wollaston Land. These are the only trees which occur even in a dwarfed state in the Barren Grounds proper. The principal trees of the spruce forest, whose northern limit is thus defined, are the white and black spruce, whose range is coextensive with the forest limits, the canoe birch, tamarack, aspen and balsam poplars, Banksian pine and balsam fir common in the southern part of the belt and terminating from south to north about in the order given. With these are associated, generally in the form of undergrowth, a number of shrubs. The tree limit on the western mountains in lati- tude 56° is about 4000 feet. The head of the Mackenzie delta is marked by islands well wooded with spruce and balsam poplar. Lower down these trees give way to willows, which continue to the Arctic shore.” South of the line noted above is a belt of conifers stretching across 1 R. McFarland, Beyond the Height-of-Land, Bull. Phil. Geog. Soc., vol. 9, rot, p. 31. 2 K. A. Preble, A Biological Investigation of the Athabaska-Mackenzie Region, North American Fauna, Bureau of Biol. Surv. No. 27, 1908. LAURENTIAN PLATEAU AND ITS OUTLIERS 571 northern North America from Hamilton Inlet to Great Slave Lake, the trees increasing in size and variety toward the south. This is the great spruce forest belt of Canada, whose characteristic growths are 1303 100° i ox 408 7s a te s a veo <3 4 iY ™ oF Roe \ ‘call Sek ee dies See A Abies balsamea divaricata »__ Abies 3D balsamea yg Picea. Fig. 228. — Map showing distribution of the dominant conifers in Canada and eastern United States. (Transeau.) spruce, fir, poplar, willow, alder, tamarack, etc., and in the Great Lake region its borders are covered with forests of deciduous hardwood.! The uniformity of level, the prevailing absence of soils, and the widespread occurrence of lakes, favor uniform life conditions through- 1 R. M. Bell, The Geographical Distribution of Forest Trees in Canada, Scottish Geog. Mag., vol. 13, 1807, pp. 281-206. 572 FOREST PHYSIOGRAPHY out the region, a fact well shown by the widespread distribution of the fur industry, which is also favored by the equally widespread forest growth, while the network of navigable streams and lakes makes com- munication relatively easy as unsettled countries go. = SUPERIOR HIGHLANDS After consideration of the topographic, drainage, and soil conditions of the Laurentian area of Canada the features of the Superior High- lands may be easily appreciated, for this physiographic province is but a part, strictly speaking, of the great Laurentian area of Canada. Like the latter, its sky line is distinctly even and gives little hint of the moun- \ Q ALS ZA Oo aN SOs = \ Seale of Miles 0 yy 1 2 Fig, 229. —Typical drainage irregularities in the Lake Superior Highlands. (U.S. Geol. Surv.) >: tainous structures that prevail almost everywhere within it. Its prin- cipal topographic feature is an uplifted and dissected plain of erosion which has been glaciated and hence has many secondary features due to ice erosion. It lies neither in the region of pronounced glacial aggra- dation nor in that of intense glacial denudation; hence those forms that are of glacial origin are due in some cases to ice scour, in others to ice accumulation. A certain amount of glacial detritus occurs here and there; in other localities the surface is swept practically clean by gla- LAURENTIAN PLATEAU AND ITS OUTLIERS 573 cial erosion. The glacial material is irregularly disposed in character- istic fashion and blocks the drainage to such an extent that lakes and ponds occur in large numbers. The most common type of depression 1e) ‘I | aay { NORTH : — CENTR . ae Q?. A "is a Seale e tC) 60 100 Miles Fig. 230. — The heavy broken line is the boundary between the Superior Highlands on the north and the Prairie Plains on the south. It is, however, more prominent as a geologic boundary than as a topo- graphic boundary. Note the connection at the western end of Lake Superior between the Superior Highlands and the Laurentian Plateau of Canada. The special districts indicated are portions of the region described in the text. is in the form of a small basin partly grown up to bushes and grasses, a swampy tract called a ‘“muskeg.”’ In describing the detailed features of the area we shall select three tracts that have been well explored and that are representative of the whole area. In examining them it is well to remember a number of general features. The region consists of crystalline rock, — gneiss, granite, and schist as well as slate, dolomite, sandstone, limestone, and other types. The crystalline rock predominates throughout the greater part of the area; the sedimentary rock occurs chiefly upon the borders. The crystalline rock dips at so high an angle that it is often vertical, thus exposing rock types of different hardnesses side by side in narrow belts. The result is that a valley carved in the soft rock may not be very deep and yet may have nearly vertical sides, so that the country 574 FOREST PHYSIOGRAPHY often has a very rugged appearance in detail that is the result of recent erosion and yet not lose on the interstream areas that plateau quality that is the result of prolonged erosion in early (pre-Cambrian) and later (Jurassic-Cretaceous) geologic time. The structure of the sedimentary rocks (Paleozoic) is in sharp contrast to that of the crystallines. The former lie in comparatively flat attitudes, dipping only slightly out- ward (southward), Fig. 232. In many places outliers of these flat to gently inclined strata occur beyond the main border of the sedimen- taries and in such instances cap the vertical crystallines and accentuate the plateau quality of the region. Their nearly vertical cliffed borders add also to the steepness of the valley sides. - Almost the entire zone of decomposed rocks formed in preglacial time in the Superior Highlands was removed by glacial erosion; in- deed glacial truncation was so pronounced as seriously to reduce the amount of available iron ore in the Lake Superior region. In many places the ice cut deeper into the soft ore bodies than into the adjacent harder rocks and thus produced subordinate valleys in sympathy with the ore deposits, a feature that is well illustrated in the Mesabi district. The soft ores were so comminuted that they are not easily distinguish- able in the drift, but the hard ores occur in the glacial drift farther southeast and indeed in the whole region so plentifully that it is clear that a great portion was swept away by the ice.! REPRESENTATIVE DISTRICTS CRYSTAL FALLS DISTRICT 2 The Crystal Falls district in the Superior Highlands is a somewhat rolling plain sloping gently downward to the southeast at an average rate of a little less than 20 feet per mile. Its average elevation is 1200 to 1300 feet above the sea. This plain is formed in part upon soft and gently inclined sandstones (Upper Cambrian) and in part upon crystalline and much harder and more highly tilted rocks (pre-Cam- brian) of varying characters, though it everywhere maintains a very uniform slope regardless of the underlying rock formations. The minor topographic features of the plain have a large variety of form and detail yet very small relief. There are no commanding eminences and the sky line is generally even. Portions of the border of the mountain 1 C. R. Van Hise, The Iron Ore Deposits of the Lake Superior Region, 21st Ann. Rept. U. S. Geol. Surv., 1899-1900, pt. 3, pp. 333-336. 2 H. L. Smythe, The Crystal Falls Iron-bearing District of Michigan, Mon. U. S. Geol. Surv. No. 36, 1899, pp. 331-333: ‘ha are composed of crystalline rock, and here the sur- ie is dotted with rocky knobs generally elongated east lid west, which is the direction both of the gneissic folia- ‘in and of the ice movement. The elevations rise 5, 10, _and in some cases 60 feet above the intervening de- essions and have steep smooth walls. The slight relief ‘| the district is indicated by the fact that the more ‘ominent elevations are deposits of modified drift. Only casionally are they due to small rock masses of harder aterial, as Michigamme Mountain, which reaches an ‘byation of 200 to 300 feet above the general level, an \pvation sufficiently great in a region of very slight relief | result in the application of the name “mountain”’ to nat would generally be regarded as a very insignificant pographic feature. The details of the topography are /imarily glacial in origin and only secondarily a response » geologic structure. The almost innumerable muskegs or basin-like depres- ns are the work of either irregular glacial cutting or ‘|ling. Where softer rock occurs the ice gouged out a \spression; where the rock is exceptionally resistant a aob or ridge is the result. Hills of glacial drift are ‘attered about the surface in large numbers. A glacial oil is also developed over a large part of the tract but hs been entirely washed away and the rock surface laid are on many steep slopes where it was originally thick ‘ad where fires and lumbering operations have removed de forest cover and allowed soil erosion to take place. ‘he structural domes of crystalline rock have a character- ‘tic topographic expression, their margins being fre- }uently abrupt and in places marked for considerable ‘istances by scarp-like slopes in the granites where ver- cal contacts with softer formations occur. These broad jones of harder rock are separated by broad and slightly pwer lying plains, in many of which a valley character » still distinguishable in spite of extensive glacial modi- cation. The existing drainage follows in a general way jhe older (preglacial) valleys, although the relation is ftentimes much confused in its details since glacial drift ow lies irregularly disposed on the floors of the pre- lacial valleys. LAURENTIAN PLATEAU AND ITS OUTLIERS 575 Kad Kse FAULT LINE 7 Mountains Ss Oy a. 344° Wp 8 AS 2% % SS Bq x Kd, diabase; kp, porphyry; ksc, sandstone; (Irving, U. S. Geol. Surv.) Fig. 231. — Deformation of strata near Porcupine Mountain, northern Michigan, in the Superior Highlands Province. Length of section, 15 miles. sandstone, conglomerate; kss, sandstone, shale. FOREST PHYSIOGRAPHY 576 (‘Aung "JO9H ‘ost \\ ‘ULWIpIa\\) ‘spuelysiy Jorredng aq} Jo Japsoq ulaY NOS 94} Jo AYdvisodo} puv sinjonys — ‘zz “BIT LAURENTIAN PLATEAU AND ITS OUTLIERS 594 SOUTHERN DISTRICT The southern portion of the Superior Highlands (central-northern Wisconsin) con- sists of a main upland, an uplifted and dis- sected peneplain with a very even summit below which the Wisconsin River and its ‘tributaries have eroded their valleys and above which project a few isolated hills and ridges. The plain gradually descends south- ward and finally diverges; one plane runs under the Potsdam sandstone which forms the northward margin of the Paleozoic de- : 1050 1000 900 Outwash Plain of Wisconsin Drift (Weidman, Wisc. Geol. Surv.) GRAND RAPIDS Wisconsin CRYSTALLINE DISTRICT & CS f Pre Cambrian posits of the region and another extends across the Paleozoic formations. In the vicinity of Grand Rapids, Wisconsin, and approximately at the border of the sand- stone district, 1000 feet above the sea, the valley bottom of the Wisconsin River prac- tically coincides with the level of the lower or pre-Cambrian plain. The pre-Cambrian and the main crystalline rocks are very much folded and crumpled and the dips are at various angles and frequently almost ver- ticalbige 223 The structures are such as are associated, in the early stages of a first topographic cycle, with topographic features of moun- tainous proportions and in their present planed condition indicate that widespread peneplanation has taken place. The strati- graphic conditions show that this peneplain was of pre-Cambrian age and was buried by Paleozoic sediments to be reéxposed by the long interval of erosion between the Creta- ceous and the Quaternary. During the Jurassic-Cretaceous cycle of erosion the mar- ginal deposits were in part stripped from the pre-Cambrian peneplain and in part “Pre Cambrian OLD DRIFT mS < SEA LEVEL a - Potsdam Sandstone Hemlock Cr, 900 FEET ABOVE SEA LEVEL PITTSVILLE Yellow River SANDSTONE DISTRICT eS oy SN) & 000T BRE = 0008 a — 0008: its} oury 408, 0L £4 ‘Sao TWAOPSULVITITAL “SSUY ASN geno vag Mo Omar ee Oa iW JONWY ooo Big IDA, OPA 000% “Bag “HF 0( Fog 43 OST esti A ‘WT 082 Buoy TIP{I}800g 1vyuaz800q how, 37 uospnyy Hives Vor LI 5 ANelalealaven NOSQNH 3 9}0N *Sod4} Yo01 sassaidxo Zurpeys pue SUO!CAV[a Ssaidx9 sInojUOD — “Lz “B17 NEWER APPALACHIANS 681 tian area of Canada, the Adirondacks, and by an exceptionally high and rugged portion of the Appalachian Plateaus, the Catskill Mountains. The northern district of the Newer Appalachians is also unlike the other districts in that the mountains upon its eastern border consist partly of metamorphosed rock, the schists of the Taconic and Mount Greylock ranges. The Newer Appalachians have a very restricted development in northwestern New Jersey, but shortly after entering New York state and specifically in the Walkill and Middle Hudson valleys the Hudson River shales thicken greatly and the whole belt has a notably broader development. Between the Highlands of the Hudson and the Catskills a broad valley lowland has been formed which continues northward along the eastern border of the state to a point between Hudson and Rensselaer, where the southern outliers of the Taconic mountains begin. From this point northward the relatively soft Hudson River shales are restricted to a narrower belt and the mountain feature of the Newer Appalachians becomes more prominent. The principal topographic fea- ture of the district is the Taconic Range. Its geographic position and subdivisions are shown on the map, Figs. 273 and 274. The broadest development of the mountain group of which it forms a part is between the Hudson and the Hoosic valleys, roughly on the parallel of Troy. At this point the Newer Appalachians consist of the following members named in. order from west to east: (1) the Hudson Valley, (2) the Rensselaer Plateau, (3) the Little Hoosic Valley, (4) the Taconic Range, including the Mount Greylock spur of the main Taconic Range, and (5) the upper portions of both the Hoosic and Housatonic valleys. From this point northward the Taconic mountains extend as a narrower range as far as northern Vermont, where they terminate. The total length of the range is about 200 miles, its width is from 5 to ro miles. Its course is somewhat serpentine, with a north-northwest trend near Great Barrington, north-northeast to Dorset, and similar turns farther north.? Many of the forms of the Taconic region are directly or indirectly 1 The name ‘‘ Berkshire Hills’’ is so commonly employed to designate portions of the region here discussed that a word as to its usage is in point. Dale (The Rensselaer Grit Plateau in New York, 13th Ann. Rept. U. S. Geol. Surv., pt. 2, 1891-92, p. 297) implies a restriction of the term to the Taconic Range and-the Greylock offshoot or the rugged country between the headwaters of the Hoosic and Housatonic rivers on the east and the western foot of the Ta- conic Range. Emerson’s usage (Holyoke Folio U. S. Geol. Surv. No. 50, 1898, p. 1) corre- sponds to that of the inhabitants of the region and includes all the hills and mountains of Berkshire County, Massachusetts. In this usage the Hoosac Mountains, the southern continu- ation of the Green Mountains of Vermont, or what is called in Massachusetts the western edge of the Green Mountains Plateau, are included with the Taconics in the term “‘ Berkshire Hills.” 682 FOREST PHYSIOGRAPHY related to deformations acquired in three periods of folding: (1) at the close of the Cambrian, affecting the central portion of the area; (2) at the close of the Ordovician, with more far-reaching effects; and (3) in post-Silurian time (Devonian or Carboniferous)... The general struc- tural features of the Taconic region are a succession of major and minor folds, as shown in the accompanying illustration, Fig. 273. These cor- respond approximately to the trend of the ranges. The general broad aspect of the structure of the Taconic Range is that of a synclinorium, in contrast to the geanticlinal character of the Green Mountains. The valleys are generally developed upon the softer limestone, the hills upon the harder schist.’ The easterly dipping cleavage of the schist determines the char- acter of the eastern slope of the hills. There is a marked arrange- ment of the drainage along north-south or longitudinal lines both within the range and in the bordering valleys. There are 5 main transverse valleys—the valley. of the Hoosic, the Mettawwe, the wide valley of the Walloomsac, the Battenkill, and the valley of the Castleton. Of these the Hoosic and the Walloomsac are deeply in- trenched in wide valleys and the Taconic Range is greatly dissected in their vicinity.’ The main forms of the Taconic mountains are narrow ridges of harder rock separated by valleys developed upon softer rock. There is con- siderable variation in the lengths and forms of the ridges. Some are long, others short. The longer ones in many cases sag gently toward the center in response to variations in the axial pitch of the anticlines, or to the exposure of softer limestones in the heart of the anticlines. The shorter ones are in some cases roughly pyramidal in outline, irregular spurs with amphitheater-like hollows between them. There is also a small number of plateau-like masses. Few cliffs occur though some of these attain heights of 1000 feet. The rugged hilltops and ridge tops of schist thinly veneered with soil have few agricultural resources and constitute areas of ultimate forest land. In general the valleys are deeply covered with drift and alluvium derived in large part from the underlying limestone. They are fertile and have an important agricultural population. The Rensselaer Plateau lies between the Taconic mountains and the Hudson River in Rensselaer County, New York, and extends northward in scattered remnants toward Castleton, Vermont. Its structural fea- 1 T. N. Dale, Taconic Physiography, Bull. U. S. Geol. Surv. No. 272, 1905, p. 48. 2 Idem, p. 20. 3 Idem, pp. 20-01. NEWER APPALACHIANS 683 tures and geographic position ally it with the Taconic mountains on the east. It rises from 700 to 1200 feet above the adjacent valleys and from 1400 to 2000 feet above sea level. Its structure is broadly syn- clinal, its rocks massive, and its relatively flat surface is the product of base-leveling. The lakes that dot its surface are chiefly due to irregular glacial erosion. It bears little good soil and in this respect is in marked contrast to the fertile Hudson Valley on the west, the Berlin and Berkshire valleys on the east, and even the Berkshire Hills. It was once thickly timbered, but has since been deforested, and is to-day an unattractive region with relatively steep slopes on the east, north, and south, and with a general westerly inclination. The extreme northern end of the Newer Appalachians is the St. Law- rence Valley, which drains out to the northeast longitudinally like the Coosa on the southwest. In this respect the extremities are in contrast to the greater part of the province, which is drained only by tribu- taries, the master streams crossing the mountain ridges and valleys alike at right angles. The limestones on the northern side of the St. Lawrence Valley have been weathered down into a lowland exten- sively covered with glacial detritus and estuarine deposits formed when the land stood lower than now and the sea washed the foot of strand lines now elevated 1500 feet above sea level. The lowland strips be- come narrower toward the northeast and finally disappear before reach- ing the mouth of the Saguenay. Beyond this point the bay of the St. Lawrence occupies the entire breadth of the lowland. TREE GROWTH The various sections of the Newer Appalachians display quite dif- ferent types of tree growth. The rich limestone valleys which form so large a portion of the southern district were originally covered with an excellent growth of hardwoods interspersed with open parks or natural prairies where the limestone is most fissured and porous and therefore most dry. At present the limestone valleys are extensively cleared owing to the exceptional fertility of their soils and the abundant supply of timber available on the uncultivated ridges of the province. Like- wise the lowland portions of the northern district are cleared and farmed. The zigzag ridges of the central district, on the other hand, are still for the most part tree covered. Their bordering slopes are quite too steep and their summit areas are quite too small to tempt the farmer. They constitute ultimate forest land of great value when kept in trees, and of little value, even as sheep or goat pastures, when cleared of their 684 FOREST PHYSIOGRAPHY timber. The drier, sandier ridges formed on resistant sandstone, such as the Pocono, are marked by stunted growths of scrub pine and red and black oak; the more fertile valleys underlain by limestone have a heavy native growth of walnut, blue ash, etc. The best growth is found on the highest ridges of eastern West Virginia where the rainfall is notably greater than elsewhere in the province. 93° 89° 85° 81) os el RP nee C f the 0 Wools AAS | anemic > a / / . Qe nI0 wey ec HANDELEUR Ne sy is. from 93° Greenwich _ 89° t fee HS Si SP AS Sas: SCALH OF TINTS ~ E Oo > fc) 2 3 D i=) fe} 3 s > te ES = val Plate 1V. —Physiographic Map of the United States. Scale of Miles 7s, = es, 4 ss. = B _ - ie -- ~ om cone PLATE IV ae INES erie ESP tke 4 ae pe ca es ot esau CHAPTER XXXII APPALACHIAN PLATEAUS THE Appalachian Plateaus consist of a number of subdivisions of the first rank: the southern or Cumberland district; the central district in- cluding the so-called “mountains” of eastern Kentucky and West Virginia; the northern district, the Allegheny Plateau of western Penn- sylvania and southern New York; and the extreme northeastern por- tion —a special category— including the Catskill Mountains. The Catskills and the plateau of West Virginia and Kentucky are the loftiest members of the series and have truly mountainous relief and propor- tions. The Allegheny Plateau has been dissected into a systemless maze of spurs by the almost infinite branches of the deeply intrenched streams. The high and rugged eastern border region of the Allegheny Plateau is also known as the Allegheny Mountains, a term which in- cludes not only the Allegheny Front but also the rough uplands imme- diately west of it, the northern representatives of the Cumberland Plateau.!' Like the Cumberland Plateau the upper altitudes of the region represent an old erosion level, but dissection has progressed to the point where practically no flat land exists on the divides as an inherit- ance from an erosion cycle long closed. The tremendous resistance of the thick border rock of the Cumberland Plateau and the lower altitude have practically preserved it from destruction and flat uplands still persist. NORTHERN DISTRICT The northern district of the Appalachian Plateaus consists of the plateaus of western and northern Pennsylvania and southern New York and include such specialized tracts as the Pocono Plateau in northeast- ern Pennsylvania and the Catskill Mountains in east-central New York. The western border of the district is a well-developed scarp that swings southwestward along the southern shore of Lake Erie, finally disappearing in central Ohio. At Cleveland it stands out as a ragged scarp several hundred feet high, between which and the lake shore there occurs but a narrow strip of lowland. Farther southwest the 1 See the Bedford quadrangle, Penn., U. S. Geol. Surv. 685 686 FOREST PHYSIOGRAPHY province has a less definite border than elsewhere except at the extreme south, where it descends gradually and dips beneath the Gulf Coastal Plain. - Its northern border is a low ragged northward-facing escarpment (Fig. 275) stretching eastward from Lake Erie to the Hudson across cen- tral New York. The eastern end of this escarpment is formed by the Helderberg Mountains, where the extremely hard Helderberg limestone outcrops in a stratum about 500 feet thick. Farther west, in the Finger Lake district, the escarpment makes a bend southward, a change in trend due to the great thickness (1000 feet) of the soft and Fig. 275. — North-south section across the northern edge of the Appalachian Plateaus, Chemung River to Glenwood, N. Y._ Vertical scale 5 times the horizontal. (Tarr, U.S. Geol. Surv.) easily eroded Salina shales. At this point too the resistant Helderberg limestone is only go feet thick. Farther west the shales thin out again, the sandstones and limestones become thicker, and the escarpment again swings back to a more northerly position. The present position of the escarpment is in a geologic sense merely temporary, for the sapping of the hard layers which are responsible for its prominence is going on now just as in the past. By the same reasoning it was once farther north than now, and has been steadily pushed southward, so that this part of the Appalachian Plateaus province, like so many other por- tions with scarped margins, is suffering a reduction in area by the extension of the lowlands about it. An interesting consequence of this process of escarpment retreat is progressive stream capture exhibited in many forms. The drainage of the plateau is southward almost from the very edge. The shorter, steeper, and more powerful northward-flowing streams are cutting back into the drainage systems of the Susquehanna and the Allegheny. In places they have diverted one tributary after another until almost the entire headwater systems of individual tributaries have been deflected to northerly courses. The junction of the deflected and the deflecting streams is marked by a sharp turn, so that the two stand in a curious relation desig- nated as barbed drainage. From the extent to which this feature is developed conclusions may be drawn as to the former position of the plateau margin in recent time. An example is West River above the head of Canandaigua Lake. The southward-flowing plateau streams whose headwaters have been captured are diminished in volume and in many cases their head- waters flow as tiny brooks in broad valleys. Indeed the upper portions of some valleys on the plateau margin are without a living stream. The eastern margin of the plateau of southern New York is also being pushed back rapidly by the short precipitous tributaries of the Hudson that descend the great scarp on the eastern aspect of the Catskills. Stream capture is here both vigorous and general. In places it means merely the westward retreat of the escarpment without sudden and great changes in the courses of the streams; at other places the valley sides are broken down by lateral attack, as in the APPALACHIAN PLATEAUS 687 escarpment of central New York, and barbed drainage relations developed. The best-known case is that of the eastward-flowing Kaaterskill, which has captured the headwaters of west- ward- and northward-flowing Scoharie Creek and turned down a 5-mile course waters which formerly flowed 50 miles to the Mohawk and down the Hudson to reach the same point. The plateau of northern Pennsylvania and southern New York con- sists of a broad elevated region so extensively and deeply dissected that only small remnants exist here and there of what was once a fairly even surface! The Cretaceous peneplain which by uplift became a plateau has by that erosion which is dependent upon uplift become a dissected plateau. Above its general level stand distinctly higher ridges with comparatively flat tops. The degree of dissection of this portion of the Appalachian Plateaus is so great that the features of the central district are repeated in kind though not in degree. The lesser elevation of the northern district has ~ resulted in shallower valleys and a less mountainous aspect than occurs in eastern Kentucky, but the relief is decidedly rugged. There is the same kind of dependence upon the valley ways in both the newer and the older systems of transportation. “Wherever the surface is underlain by the same set of strata, it is cut into hills and val- leys of the same general style. One valley can not be called the counterpart of another, nor are the hills and uplands always alike, yet the type of topography is the same. A view from the top of any well-exposed upland gives a good idea of what the country is like. Below the upland lies a valley as variable in the nature of its slopes as it is irregular in its course. Here steep walls rise from thé stream on both sides. There a sharp descent on one side is faced by a long gentle slope on the other. Numerous ravines, some short, some long, some deep, some shallow, are occupied by the smaller streams which flow in from either side. As far as the eye can see in all directions the uplands stretch away in a broad, undulating tableland, unbroken by ridges, but on every hand bearing the deep scars of a multitude of valleys and ravines.” 2 “Over the entire area the streams branch again and again, until there is hardly a square mile into which one or more has not worked its way . . . the surface is that of a well-dissected plateau, varying from place to place, both in elevation and in surface detail, yet everywhere preserving the general feature of more or less rugged relief produced by the trenching valleys of innumerable streams.’’3 The topographic studies thus far made in Pennsylvania seem to show that three erosion levels can be identified. The first is the level of the ridge and hilltops in northern Pennsylvania, a feature equally well shown on the summits of the zigzag ridges of central Pennsylvania and northern New Jersey. This is the Cretaceous peneplain, and was de- veloped widely upon rocks of diverse structure and resistance. The 1M. R. Campbell, Geological Development of Northern Pennsylvania and Southern New York, Bull. Geol. Soc. Am., vol. 14, 1903, pp. 277-206. 2 W. S. Tower, Regional and Economic Geography of Pennsylvania, pt. 1, Physiography, The Central Province, The Plateau Province, Bull. Geog. Soc. Phil., vol. 4, 1906, p. 30. 3 Idem, p. 36. 688 FOREST PHYSIOGRAPHY second peneplain was early Tertiary and is known as the Harrisburg peneplain. It was developed upon the Chemung rocks of the northern part of Pennsylvania, and is also well shown in the Monongahela Valley on the Brownsville, Missiontown, Connellsville, and Union Town topo- graphic sheets, where the surface is a very gently undulating plain which in a distant view has an almost horizontal sky line at 1250 feet. The Harrisburg peneplain is best developed east of Harrisburg, where it aR arrisburg \ eA SS teed } DS Philadelphiay / Sm 7 oes oe ees i Fig. 276. — Warped surface of the early Tertiary (Harrisburg) peneplain of the central Appalachians. (Campbell.) is at an altitude of 500 feet. It rises steadily upstream to about 800 feet in the vicinity of Sunbury, and to 1200 and 1300 feet at Pittston. Its present warped attitude is shown in Fig. 276. In the Schuylkill the rocks are considerably disturbed and consist of a heterogeneous mass of shales; yet the hilltops are very regular indeed, with occasional monadnocks rising as high as 800 feet above the general 500-foot level. On the northwestern side of the Shenandoah Val- ley is a region of low flat-topped hills which appears like a great plain trenched by many small valleys. In the Potomac Valley (Hancock quadrangle) the Harrisburg peneplain is from 600 to 700 feet above the sea in the southeastern corner and about 800 feet in the northwestern corner. The geologic structure consists of broad open folds, with many minor wrinkles, and the planes of stratification are generally inclined. In spite of these structural complications the peneplain represented by the hilltops cuts across the beds whatever their angles of inclination. The third topographic level is that formed in late Tertiary time; it is known as the Somerville or the Worthington plain. In New Jersey it was developed on the rocks of the Kittatinny Valley and on the wide outcrop of Triassic rocks which form the lowland belt across New APPALACHIAN PLATEAUS 689 Jersey and Pennsylvania, Fig. 276. It is, however, of exceedingly local development, and along the Susquehanna Valley as at Harrisburg it stands at an altitude of 400 feet, at Lancaster at 350 feet, and on the Potomac River near Harpers Ferry at 500 feet. It represents a partial cycle only, and nowhere was developed extensively across rocks of differ- ent hardnesses, but was etched out upon the softer formations only. WORTHINGTON PENEPLAIN PARKER STRATH SUBSTAGE SUBSTAGE WISCONSIN TERRACE B WORTHINGTON PENEPLAIN Fig. 277. — The upper section illustrates the terraces of the Ohio Valley, the lower the terraces of the Allegheny Valley. (Top. and Geol. Surv. of Penn.) In the uplift of the third and lowest plain to the present level there was one main halt which permitted the development of broad valleys. Fragments of these valleys now occur as benches along the valley mar- gins. All these features as well as those related to the glaciation of the region on the north are shown in Fig. 277. Among the more prominent effects of glaciation was the development of an extensive system of abandoned channels whose origin was long in doubt. They are of widespread occurrence in western Pennsylvania, West Virginia, and eastern Kentucky. Detailed surveys and studies in recent years have at last supplied a basis for an acceptable explana- tion! The abandoned channels appear to have been upbuilt by the streams during a period of stream aggradation associated with the waning stages of glaciation. All the southward-flowing streams heading in the glaciated country were so abundantly supplied with material that they aggraded their valley floors. The northward-flowing streams join- ing the southward-flowing aggrading streams were therefore compelled to aggrade their courses to the same level. In many cases they developed 1 E. W. Shaw, High Terraces and Abandoned Valleys in Western Pennsylvania, Jour. Geol., vol. 19, I91I, pp. 140-156. 690 FOREST PHYSIOGRAPHY courses to one side or the other of the older courses, crossed low points in former upland spurs, and now exhibit most striking anomalies with respect to earlier channels. Typical con- a ditions are represented along the Monon- g pitishime gahela and its tributaries, as shown in —| : eect ae POCONO PLATEAU MC ~, =e Greensburg Sf OOo we 2 —~ The Pocono Plateau, a separate division of the northern district, lies in the north- eastern corner of Pennsylvania, almost completely separated from the rest of the ‘i plateau areas of that state by the deep i synclinal trough of the Wyoming Val- ley. The Pocono Plateau covers Monroe, Pike, Wayne, and eastern Carbon counties, merges southwestward into the ridge area of the anthracite coal region, and extends east and north to become the Catskill Pla- teau of New York state. Almost the whole of it is underlain by the broad Fig. 278. — Present and pre-Pleistocene Strata of nearly horizontal hard sandstone courses of Monongahela (left) and and conglomerate at the bottom of the Youghiogheny rivers. (Top. and ee Ceclesiy ior Penn) Coal Measures series. The southern edge of the plateau is often known as Pocono Mountain and presents a close analogy to the Alle- gheny front, both in origin and in character. It is an erosional escarpment, tooo feet high, with a step-like ascent resulting from the horizontal position of the strata. The plateau back of the boundary escarpment (Pocono Mountain) is a nearly level upland wilderness, known to the early settlers as the “Great Beech Woods” and the “Shades of Death.”” Over most of its extent it stands from 1400 to 1800 feet above the sea. Below this level the streams have cut valleys from 100 to 200 feet deep, and an occasional knob rises 200 or more feet above the upland. It is often described as one of the wildest parts of the state, “‘a wilderness of forest and swamp,” but it is made picturesque by the numerous lakes and cascading streams that are the result of the glacial action to which the region has been subjected. BRC NSVILLE TE ONNELLSVILLE “Ae el: i ©Uniontown UNIONTO 1 W. S. Tower, Regional and Economic Geography of Pennsylvania (Physiography), Bull. Geog. Soc. Phil., vol. 4, 1906, pp. 216-217. APPALACHIAN PLATEAUS 601 CATSKILL MOUNTAINS The Catskill Mountains stand upon the northeastern border of the Appalachian Plateaus as a conspicuous group of ridges and peaks of mountainous proportions. Their summits reach to heights almost double those of the adjacent plateau. Some of the principal peaks are Slide Mountain (4220 feet), Hunter Mountain (4052), Black Dome (4000), Windham High Peak (3809), etc. From the upper portions of the Hudson Valley their whole elevation may be seen in a single view, from which point they appear to have imposing form and height. The blue haze that generally hangs over them— a feature of rare beauty in autumn weather — lends to their height a majesty and to their out- lines a softness which in a distant view blend to form one of the most charming sights of the Atlantic slope. The structure of the Catskills is as simple as that of the neighboring plateau on the west. The strata lie almost flat, with slight dips to the west, northwest, and southwest in various places. Anticlinal and synclinal structures are practically absent, and even when present they trend not with the ranges of the Catskills but at right angles to them, showing no relation to the present topographic forms. Shale com- monly outcrops on the lower slopes of the valleys, but sandstones occur higher in the section, and on the summits of the principal peaks the rock is generally a conglomerate, very durable and thick. The flatness of the strata is expressed in the flat summits of the mountains, a characteristic feature and one that often interfczcs with the view, since these mountains are all heavily wooded. The tops are often of consider- able extent and are never sharp-pointed peaks as in a region of alpine forms. While the valleys among the mountains are broad and open their sides are often cliffed to a notable extent for some distance. This is due to the system of almost vertical joints, which are the principal lines of weakness along which secondary erosion and valley widening take place. Abrupt ledges are frequent and are often a source of great difficulty in ascending a peak by unusual paths. In a few places these ledges are of great height and afford splendid panoramas over the surrounding country, as at the Catskill Mountain House and Overlook Mountain. To the vertical jointing and erosion along the joints is also to be at- tributed the successive steps which are common features of the valley floors and give rise to numerous picturesque cascades. There are two main ranges in the Catskills, the line of division being marked by Esopus Creek. The southern Catskills consist of a massive central chain which bears the highest mountain in the Catskills, Slide 692 FOREST PHYSIOGRAPHY Mountain. The roughness of the topography and the unbroken forest that covers the mountain makes the penetration of this part of the Catskills very difficult. The northern ranges of the Catskills trend northwest and the principal range is about 35 miles in length. It makes a sharp curve, bending back upon itself in a sort of secondary range. A number of lateral spurs or secondary ranges trend at right angles to the main ones. Toward the south and southwest the main range falls off in long slopes and heavy spurs. It is divided into four sections by three deep gorges or “ cloves” which give access to the interior valleys. The drainage of the Catskills is chiefly to the west by tributaries of the Scoharie system. This appears to be a consequent drainage de- veloped upon the original surface at the time of uplift of the region. Base-leveling, which is so common a feature of the Appalachian Plateaus, appears to be absent here; the Catskills appear to have survived as residuals because of their original superior height and the protection of the heavy cap of horizontal conglomerate. Erosion produced important effects, however, and opened up the broad valleys that are character- istic of the region. A common feature of the valleys is the presence of a pronounced shoulder halfway up the valley slopes, a feature which appears to be related to later uplift and dissection in the Tertiary. Were the valleys filled almost to the level of the shoulder we should probably have a picture of the Catskills as they appeared at the close of the Cretaceous cycle of denudation. They would then present smooth flowing outlines without any important number of steep ledges; the latter are commonly found below the level of the shoulder, where vigor- ous erosion is now taking place. The effects of Tertiary erosion are also shown in the form of local lowlands of limited extent opened up here and there within the mountain borders. Although the Catskills were overridden by ice, signs of which are everywhere abundant, the ice appears not to have had any important effect upon the topography; rather it conformed to the broad slopes, only slightly molding them here and there by the deposition of small quantities of glacial till or by the erosion of the sharper forms.! SOILS AND VEGETATION The most uniform type of soil in the glaciated northern portion of the Allegheny Plateau is the till sheet which veneers the hills and up- lands. It is a smooth, thin, and locally stony sheet of bowlder clay. 1 For an excellent topographic description from which the above is largely derived see Arnold Guyot, On the Physical Structure and Hypsometry of the Catskill Mountain Region, Am. Jour. Sci., 3d Series, vol. 19, 1880, pp. 429-451. APPALACHIAN PLATEAUS 693 The soil is much deeper in the valleys and of more variable form and composition, especially where morainic accumulations occur side by side with fluvio-glacial material—kames, eskers, outwash plains, and the like. In both the uplands and the valleys a large proportion of the glacially derived material is from underlying shales which weather into clay and increase the fertility of the soil by increasing its water-holding capacity. The varying proportion of this soil element causes great variation in the rapidity and thoroughness of soil drainage, so that in a dry spell one part of a field may be covered with a fresh green growth while another part near by may be parched and brown.t| Swampy tracts I We \ MA O4\ D> \ We (We Fig. 279. — Distribution of morainal deposits and direction of ice movement in western New York. (Tarr, U. S. Geol. Surv.) are characterized by a fertile black muck of great value when drained. Besides these soil types are narrow strips of flood-plain deposits and fan-shaped alluvial accumulations where the hill streams descend abruptly to the main valley floors. While practically none of the primitive forest can now be found in the region, yet the uplands and valley slopes, too steep for cultivation, are in the main covered with extensive forests. The forest cover is extending naturally and encroaching on the cleared lands. The thin, stony upland soils are relatively infertile, and the extension of the forest 1 R.S. Tarr (Williams, Tarr, and Kindle), Watkins Glen-Catatonk Folio U. S. Geol. Surv. No. 169, 1900, p. 33- 694 FOREST PHYSIOGRAPHY over them would benefit the region not only from the standpoint of forest products but also from that of the agricultural interests of the valleys, which suffer from increasingly destructive floods. : CENTRAL DISTRICT The central district of the Appalachian Plateaus lies in eastern Ken- tucky and West Virginia and is the ruggedest portion of the entire prov- ince. The highest portions are from 3000 to 4000 feet above sea level and are known locally as mountains, a name they fully deserve. The eastern rim of the district in Kentucky is Pine Mountain, whose steep eastern escarpment rises from 800 to 1500 feet above the Great Valley i Cee ELE So me, LEE LE De es 1% 9 1 2 3 4 5 Scale of Miles Fig. 280.— Maturely dissected Allegheny Plateau in West Virginia. The large depression on the right is the valley of the Kanawha at Charleston. Note the prevalence of slopes and the absence of flat land. on the east and has but one water gap in 150 miles. The Kentucky “Mountains” are structurally a part of the Cumberland Plateau and formerly had a flattish summit; the original nearly flat surface has been so greatly dissected however that but few remnants remain to in- dicate the former level. These, however, show a remarkable uniformity of elevation on northeast-southwest lines, so that the eastern escarp- ment of the district has an almost perfectly straight sky line. The western slopes of the district are in sharp contrast to the eastern. There is a bordering escarpment, but it is highly irregular, the streams having carved valleys far back into the elevated portions of the upland, leaving long narrow spurs running out toward the west. The valleys are steep and gorge-like; the main streams such as the Kanawha occupy canyons; flat land is seldom found to any extent either upon the hill sum- mits or on the valley floors. It isa hill-and-valley country, Fig. 280, where so little flat land occurs that the water everywhere falls upon a slope and the run-off is heavy; floods due to spring rains and the melting of late winter snows are common and rise to great heights in the confined valleys, destroying roads and bridges, washing away the valley soils or covering the soils of the narrow flats or flood plains with heavy deposits of coarse waste. Almost all the roads and trails follow the valleys, to which the railroad lines themselves are strictly confined. 1 R. S. Tarr (Williams, Tarr, and Kindle), Watkins Glen-Catatonk Folio U. S. Geol. Surv No. 169, 1909, p. 33- APPALACHIAN PLATEAUS 695 Travel by any means is often suspended during time of high water. The whole section is a great forested wilderness whose resources of coal and timber constitute its chief wealth, resources slow in development because of the almost insuperable topographic obstacles to transporta- tion. Even the railways on the margin of the country have been built since 1880. Magnificent forests once covered the central district of the plateau region. They are still untouched in the remoter localities. Oak, wal- nut, poplar, chestnut, maple, ash, and tulip trees grow to great size. The making of staves of white oak is a considerable industry among an isolated and backward mountaineer folk. Since the roughness of the country limits the railways to the principal streams, the lesser water- ways are almost everywhere utilized for rafting both timber and lumber to.the railroads and the lowlands. In many sections remoteness from transportation lines forbids any attempt at forest exploitation. As in the southern Appalachians a wasteful system of agriculture is practiced. Hillside farms are cleared of the finest timber by combined girdling and later burning, then cultivated a few years and abandoned for a new site. The abandoned clearing grows up to useless brush or is deeply gullied and the thin soil washed away.1 SOUTHERN DISTRICT CUMBERLAND PLATEAU, WALDEN RIDGE, LOOKOUT MOUNTAINS, ETC., AND THE HIGHLAND RIM A first inspection of the flat-topped plateaus of the southern district leads one to the conclusion that the flatness is a function of the struc- ture, for the strata appear to have a roughly horizontal attitude. A closer examination, however, reveals the fact that the plateau surface does not generally coincide with a particular stratum however resistant it may be; the surface is found to be composed of very soft shale as 1 For extremely interesting and accurate descriptions of both the physical geography of the region and the Kentucky mountaineers see the writings of John Fox, Jr., as for example Hell-fer-Sartin, Blue Grass and Rhododendron, and The Trail of the Lonesome Pine. On Horseback to Kingdom Come, Scribner’s Mag., vol. 48, No. 2, ro10, p. 175, discusses later in- dustrial development. The best scientific description of the country and the people is by E. C. Semple, The Anglo-Saxons of the Kentucky Mountains, A Study in Anthropogeography, Geog. Jour., vol. 17, t901, pp. 588-623. In Theodore Roosevelt’s The Winning of the West (The Spread of English-speaking Peoples), vol. 1, ed. of 1905, pp. 146-147, is a fascinating de- scription of the forest of pioneer days. For a good topographic description of portions of eastern Kentucky see Ky. Geol. Surv., vol. 5, n. s., 1880. For a discussion of the natural water routes of the Kentucky mountains see N. S. Shaler, The Transportation Routes of Kentucky and their Relation to the Economic Resources of the Commonwealth, Ky. Geol Surv., vol. 3, pt. 5, 2d series, 1877. 696 FOREST PHYSIOGRAPHY well as of very hard sandstone. In short, the surfaces of the various plateaus truncate hard and soft beds and are much more nearly hori- zontal than the strata upon which they have been developed. Following the principles we have already applied so frequently in the study of the physiography of the United States, we shall conclude that the surfaces of Walden Plateau, Cumberland Plateau, Lookout Mountain, Sand Mountain, etc., are portions of an uplifted peneplain in process of more or less rapid dissection. Along certain lines narrow anticlinal folds had developed during the period of structural deforma- tion, with broad synclines between. The projection of the anticlines above the level of the peneplain resulted in their erosion and the ex- posure of softer underlying beds lying in the heart of the anticlines. When later uplift occurred, opportunity was afforded for the dissection of the softer exposed beds. The result has been that the synclinal basins of an earlier period have been converted into the mountains and plateaus of the present period. One of the most important departures from the plateau topography is known as the Sequatchie Valley, which separates Walden Plateau from Cumberland Plateau. It lies parallel with the Great Appalachian Valley, has remarkable continuity and regularity of expression for over too miles, and appears to be an outlying anticlinal fold of the Appa- Jachian system of folds. The Sequatchie anticline, like the anticlines of the Newer Appalachians, has a typically unsymmetrical form, the beds dipping much more steeply on one side of the axis than on the other, and the gentler dips are upon the eastern side. Near the upper end of the Sequatchie Valley the strata have been broken by a thrust fault developed along the steep side of the arch. Walden Plateau on the east shows a distinct synclinal structure. Of similar structure is Lookout Mountain, but it is much narrower than Walden Plateau. Wills Valley, developed upon an anticline, separates the syncline of Lookout Mountain from the syncline of Sand Mountain, and Sand Mountain is in turn separated from Cumberland Plateau by the valley of Tennessee River, which has been developed in the southwestward extension of the Sequatchie anticline. Along the eastern edge of Cumberland Plateau the strata dip west- ward at a steep angle, but these dips are maintained for very short distances, usually not more than a few rods, where they change to dips that are sensibly flat, a condition maintained across the Cumberland Plateau and the Highland Rim. Though apparently horizontal the beds dip toward the southeast from 20 to 30 feet per mile. It must not be supposed that Cumberland Plateau and Walden Plateau were perfectly peneplaned. Along the western edge of Walden Plateau, the northern end of Lookout Mountain, the eastern edge of Cumberland Plateau, and at a large number of isolated points elsewhere, APPALACHIAN PLATEAUS 607 residuals in the form of isolated knobs or mesas rise from 100 to 300 feet above the general level of the plateau. In places, the residuals are composed of more resistant beds of massive conglomerate, but the resid- uals within the borders of the plateaus are composed of horizontal strata In some cases capped by a bed of conglomerate but more often composed entirely of rather soft sandstones and shales. Following the uplift of the Cretaceous peneplain there ensued a period of crustal stability sufficiently prolonged to enable the forces of erosion to develop a partial peneplain at a level from a few hundred to a thou- sand feet lower than the Cretaceous. The peneplanation accomplished during this period was chiefly upon the softer rock of the region and was so incomplete as to leave large portions of the Cretaceous peneplain standing above the early Tertiary peneplain in the form of massive unakas, as shown in Fig. 238. The early Tertiary peneplain is called the ‘Highland Rim” peneplain because the Highland Rim, so-called, between the Nashville basin and the Cumberland Plateau, is the best- preserved portion. As in the case of the well-preserved remnants of the Cretaceous peneplain, the remnants of the early Tertiary or Highland Rim peneplain have been preserved largely because of the pres- ence of resistant beds along the outer margin, but the peneplain as a whole truncates beds of widely differing degrees of resistance to erosion. The peneplain is preserved upon rocks of intermediate resistance, chiefly siliceous limestones and sandy shales, for rocks ef greater re- sistance than these were never base-leveled, and rocks of less resistance were dissected in the uplift which closed the early Tertiary cycle of erosion. The elevation of the Highland Rim peneplain is about 1ooo feet, west of the Cumberland Plateau in the Appalachian Valley it is about 1150 feet, toward the northern edge of the Chattanooga district 950 feet. Elevations above the topographic level developed on the Highland Rim are in the form of isolated residuals similar to those we have noted upon the Cumberland Plateau, as long irregular projecting spurs along the ragged west border of the Cumberland Plateau, or as massive unakas such as Cumberland and Walden plateaus themselves. On the western and southern borders of Cumberland Plateau many long spurs and isolated knobs, products of circumdenudation, project irregularly westward over the surface of the Highland Rim, breaking the continuity of the eastern portion. Between the spurs are great gorges from 800 to tooo feet deep, which, on account of their depth and narrowness, are known as “‘gulfs.” At their heads coves are found, headwater alcoves which usually contain a pocket of limestone soil that supports a better timber growth than the flat upper surfaces of the upper and lower plateaus with their thin soils formed ‘upon sandstones 693 FOREST PHYSIOGRAPHY and shales. Both the distribution of the soils and the character of the topography on the border of the Cumberland Plateau are due to the structure. : Between the hard sandstone and conglomerate capping most of the Cumberland Plateau. and the rock of the lower country about it are soft limestones and shales which are so easily eroded as to result in the sapping and undermining of the hard formations above them. It is this process which maintains the steepness of the border scarps (especially their upper portions, which are frequently vertical cliffs) and results in the sharp line of delineation between the Appalachian Plateaus on the one hand and the Appalachian Valley and Highland Rim on the other. These features are persistent, being found on the projecting portions of the plateau as well as at the heads of the coves. The “barrens” of Tennessee are developed chiefly upon shales (Waverly) which yield a white, siliceous, and unproductive soil: Scarcely more productive are the soils derived from sandstones which are thinly inhabited and have thin native forests of pine and oak. The surface of Cumberland Plateau, consisting of sandstones and shales, is covered with a thin poor soil; the Highland Rim is also far from having a pro- ductive soil, though on its western margin, where a limestone (New- man) outcrops, a soil of greater but not of high fertility occurs. In general the timber covering of the fiat plateau summits and remnants is thin, but in the coves, hollows, and gorges, where a richer soil and more abundant water supply are found, hickory, chestnut, and oak reach a good size and grow in first-class stands. Near the watercourses, pine, hemlock, and spruce find the necessary elements of their environment, but their growth is everywhere second in importance to the members of the first-named series.” LIMESTONE SOILS OF THE APPALACHIAN VALLEYS “The limestone soils are among the most extensively developed of any in the United States and occur in both broad upland and enclosed narrow valley areas. The greatest upland de- velopment is seen upon the Cumberland Plateau in eastern Tennessee and Kentucky and upon the Carboniferous formation in central Tennessee and Kentucky, northern Alabama and Georgia, and in Missouri. The valley soils are found principally in Pennsylvania, Maryland, and Virginia, and in the mountain section of eastern Tennessee and Kentucky and northern Alabama and Georgia.” 3 The limestone soils are residual in origin, being derived from the weathering in place of limestone of several ages and variable composition. This is accomplished by the removal through solution of the calcium car- bonate of the limestone. Limestone soils are remarkable for the fact 1 C. W. Hayes, Seuanee Folio U. S. Geol. Surv. No. 8, 1894, p. I. 2 A. Keith, Wartburg Folio U.S. Geol. Surv. No. 40, 1897, p. 4, col. 3, and M. R. Campbell, Standingstone Folio U. S. Geol. Surv. No. 53, 1800, pp. 4-5. 3 Soil Survey Field.Book, U.S. Bur. of Soils, 1906. ‘APPALACHIAN PLATEAUS Contour Interval.100 Feet r. (a) y)) « N \ Nea ry UG, EWE y Z GA“ = = 6, C ¢ ig ve AA CEL 699 > Lael a a 3 fo) 12) as) vo 0 bo = uM Seal rs) = o a 3 2 ra oO vo 5 ra vo ves pa cay o Ze an ae (=iie) Sv ENO) ae a roi op oy de